Mixed-state topological order and the errorfield double formulation of decoherence-induced transitions
Extracted source text ยท arXiv 2301.05687
\documentclass[aps,prb,notitlepage,superscriptaddress,nofootinbib]{revtex4-1}
\usepackage{graphicx,amssymb,amsmath,amsfonts,empheq}
\usepackage[dvipsnames]{xcolor}
\usepackage{color}
\usepackage{braket}
\usepackage{latexsym}
\usepackage{upgreek}
\usepackage{dsfont}
\usepackage{bm}
\usepackage{makecell}
\usepackage{multirow}
\usepackage{enumitem}
\usepackage[normalem]{ulem}
\usepackage{url}
\usepackage{bbold}
\usepackage{hyperref}
\hypersetup{
colorlinks = true,
linkcolor = [rgb]{0,0,0},
citecolor = [rgb]{0.13,0.55,0.13},
urlcolor = [rgb]{0.25, 0.41, 0.88}}
\usepackage{bbm}
%
\usepackage{tikz}
\usepackage{tikz-cd}
\usetikzlibrary{arrows}
\usetikzlibrary{intersections}
\usetikzlibrary{shapes.geometric}
\usetikzlibrary{decorations.pathmorphing, patterns,shapes}
\usetikzlibrary{decorations.markings}
%
\tikzset{
partial ellipse/.style args={#1:#2:#3}{
insert path={+ (#1:#3) arc (#1:#2:#3)}
}
}
\tikzset{
%
mid arrow/.style={postaction={decorate,decoration={
markings,
mark=at position .575 with {\arrow[#1]{stealth}}
}}},
near arrow/.style={postaction={decorate,decoration={
markings,
mark=at position .275 with {\arrow[#1]{stealth}}
}}},
far arrow/.style={postaction={decorate,decoration={
markings,
mark=at position .800 with {\arrow[#1]{stealth}}
}}},
}
\pgfdeclarepatternformonly{south west lines}{\pgfqpoint{-0pt}{-0pt}}{\pgfqpoint{3pt}{3pt}}{\pgfqpoint{3pt}{3pt}}{
\pgfsetlinewidth{0.4pt}
\pgfpathmoveto{\pgfqpoint{0pt}{0pt}}
\pgfpathlineto{\pgfqpoint{3pt}{3pt}}
\pgfpathmoveto{\pgfqpoint{2.8pt}{-.2pt}}
\pgfpathlineto{\pgfqpoint{3.2pt}{.2pt}}
\pgfpathmoveto{\pgfqpoint{-.2pt}{2.8pt}}
\pgfpathlineto{\pgfqpoint{.2pt}{3.2pt}}
\pgfusepath{stroke}}
%
\newcommand{\bbrakket}[2]{\mbox{$ \langle\!\langle #1 | #2 \rangle\!\rangle $}}
\newcommand{\moy}[1]{\langle #1 \rangle}
\newcommand{\kket}[1]{\mbox{$| #1 \rangle\!\rangle$}}
\newcommand{\bbra}[1]{\mbox{$\langle\!\langle #1 |$}}
\renewcommand{\bar}{\overline}
%
%
\renewcommand{\leq}{\leqslant}
\renewcommand{\geq}{\geqslant}
\renewcommand{\Re}{\operatorname{Re}}
\renewcommand{\Im}{\operatorname{Im}}
\newcommand{\rd}{\mathrm{d}}
\newcommand{\ri}{\mathrm{i}}
\newcommand{\tr}{\operatorname{tr}}
\newcommand{\Tr}{\operatorname{Tr}}
\newcommand{\sgn}{\operatorname{sgn}}
\newcommand{\nn}{\nonumber}
\newcommand{\Pf}{\operatorname{Pf}}
\newcommand{\const}{\operatorname{const}}
\newcommand{\U}{\operatorname{U}}
\renewcommand{\O}{\operatorname{O}}
\newcommand{\SU}{\operatorname{SU}}
\newcommand{\SO}{\operatorname{SO}}
\newcommand{\SL}{\operatorname{SL}}
\newcommand{\GL}{\operatorname{GL}}
\newcommand{\PSL}{\operatorname{PSL}}
\newcommand{\T}{\operatorname{\textbf{T}}}
\newcommand{\Sch}{\operatorname{Sch}}
%
\newcommand{\ii}{\mathrm{i}}
\newcommand{\dd}{\mathrm{d}}
%
\newcommand{\bbC}{\mathbb{C}}
\newcommand{\bbD}{\mathbb{D}}
\newcommand{\bbE}{\mathbb{E}}
\newcommand{\bbH}{\mathbb{H}}
\newcommand{\bbI}{\mathbb{I}}
\newcommand{\bbN}{\mathbb{N}}
\newcommand{\bbP}{\mathbb{P}}
\newcommand{\bbQ}{\mathbb{Q}}
\newcommand{\bbR}{\mathbb{R}}
\newcommand{\bbS}{\mathbb{S}}
\newcommand{\bbU}{\mathbb{U}}
\newcommand{\bbV}{\mathbb{V}}
\newcommand{\bbZ}{\mathbb{Z}}
%
\newcommand{\calA}{\mathcal{A}}
\newcommand{\calB}{\mathcal{B}}
\newcommand{\calC}{\mathcal{C}}
\newcommand{\calD}{\mathcal{D}}
\newcommand{\calE}{\mathcal{E}}
\newcommand{\calF}{\mathcal{F}}
\newcommand{\calG}{\mathcal{G}}
\newcommand{\calH}{\mathcal{H}}
\newcommand{\calI}{\mathcal{I}}
\newcommand{\calJ}{\mathcal{J}}
\newcommand{\calK}{\mathcal{K}}
\newcommand{\calL}{\mathcal{L}}
\newcommand{\calM}{\mathcal{M}}
\newcommand{\calN}{\mathcal{N}}
\newcommand{\calO}{\mathcal{O}}
\newcommand{\calP}{\mathcal{P}}
\newcommand{\calQ}{\mathcal{Q}}
\newcommand{\calR}{\mathcal{R}}
\newcommand{\calS}{\mathcal{S}}
\newcommand{\calT}{\mathcal{T}}
\newcommand{\calU}{\mathcal{U}}
\newcommand{\calV}{\mathcal{V}}
\newcommand{\calW}{\mathcal{W}}
\newcommand{\calX}{\mathcal{X}}
\newcommand{\calY}{\mathcal{Y}}
\newcommand{\calZ}{\mathcal{Z}}
%
\newcommand{\eqnref}[1]{Eq.~\eqref{#1}}
\newcommand{\figref}[1]{Fig.~\ref{#1}}
\newcommand{\appref}[1]{Appendix~\ref{#1}}
\newcommand{\tabref}[1]{Tab.\,\ref{#1}}
\newcommand{\secref}[1]{Sec~\ref{#1}}
\newcommand{\EFD}{\mathrm{EFD}}
\newcommand{\vect}[1]{{\bm{#1}}}
\newcommand{\norm}[1]{{\lVert #1\rVert}}
\newcommand{\smat}[1]{\left[\begin{smallmatrix}#1\end{smallmatrix}\right]}
\setcounter{equation}{0}
\setcounter{figure}{0}
\setcounter{table}{0}
\setcounter{page}{1}
\setcounter{section}{0}
\renewcommand{\theequation}{S\arabic{equation}}
\renewcommand{\thefigure}{S\arabic{figure}}
\renewcommand{\thesection}{S\Roman{section}}
\renewcommand{\thepage}{S\arabic{page}}
%
\newcommand{\RF}[1]{\textcolor{red}{#1}}
\newcommand{\yimu}[1]{\textcolor{blue}{#1}}
\begin{document}
\title{Supplementary Online Material for ``Mixed-state topological order and the errorfield double formulation of decoherence-induced transitions''}
\date{January 13, 2023}
\author{Yimu Bao}
\affiliation{Department of Physics, University of California, Berkeley, CA 94720, USA}
\author{Ruihua Fan}
\affiliation{Department of Physics, Harvard University, Cambridge, MA 02138, USA}
\author{Ashvin Vishwanath}
\affiliation{Department of Physics, Harvard University, Cambridge, MA 02138, USA}
\author{Ehud Altman}
\affiliation{Department of Physics, University of CA, Berkeley, CA 94720, USA}
\affiliation{Materials Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA}
\maketitle
\tableofcontents
\section{Phase transition in Toric code subject to incoherent errors}
In this section, we provide details of the phase transition in the Toric code subject to bit-flip errors.
%
In Sec.~\ref{sec:transition}, we show that the anyon condensation transition in the EFD is exactly solvable.
%
It occurs at $p_c^{(2)} = 0.178$ and is of the 2D classical Ising universality.
%
%
%
%
%
%
In Sec.~\ref{sec:TEE}, we compute the topological entanglement entropy in the EFD state and show it sharply changes from $2\log 2$ to $\log 2$ at the condensation transition.
\subsection{Critical point}\label{sec:transition}
As mentioned in the main text, the bit-flip errors in the Toric code can induce a condensation of $m\bar{m}$-anyon in the EFD at a finite error threshold.
%
The transition is exactly solvable and of 2D classical Ising universality.
%
Here, we provide details on exactly solving the transition.
To start, we consider the EFD state describing the Toric code subject to bit-flip (Pauli-X) errors
\begin{equation}
\label{eq:EFD_TC_bitflip}
\kket{\rho} = \prod_{l} \left( (1-p) + p X_l\otimes X_l \right) \ket{\Psi_0}\otimes \ket{\Psi_0^*}\, \propto e^{\mu \sum_i X_l \otimes X_l} \ket{\Psi_0}\otimes \ket{\Psi_0^*},
\end{equation}
where $p$ is the bit-flip error rate, $\mu = \tanh^{-1}[p/(1-p)]$, $l$ labels the edges in the square lattice, and $\ket{\Psi_0}$ is one of the Toric code ground states specified below.
We work in a loop picture of the Toric code ground state
\begin{align}
\ket{\Psi_0} \propto \prod_p \frac{1 + B_p}{2} \ket{+} \propto \sum_{\{g\}} \ket{g},
\end{align}
where $p$ in the subscript labels the plaquette, $\ket{+} := \prod_i \ket{+}_i$ is a product state with each qubit initialized in the $+1$ eigenstate of Pauli-X operator, i.e. $X_i \ket{+} = \ket{+}$.
%
A product of $B_p$ gives $g_z$, which is a product of the Pauli-Z operators along a closed loop.
%
When acting on $\ket{+}$, $g_z$ flips qubits into $-1$ eigenstates of $X_i$ along a loop $g$ and leads to a product state $\ket{g} := g_z \ket{+}$.
%
This allows writing the ground state in terms of closed loops $g$, represented by binary variables.
The EFD can be written in the loop picture accordingly.
%
Before applying the error channel, the double ground state is expressed as
\begin{align}
\kket{\rho_0} = \ket{\Psi_0} \otimes \ket{\Psi_0^*} = \frac{1}{\sqrt{\calZ(0)}}\sum_{g,\bar{g}} \kket{g, \bar{g}},
\end{align}
where $\kket{g, \bar{g}} := (g_z \otimes \bar{g}_z) \ket{+}\otimes \ket{+}$, $g$ and $\bar{g}$ label two independent loop configurations in the ket and bra copy, respectively.
%
The error channel assigns a weight in the summation of loops and leads to
\begin{align}
\kket{\rho} = \frac{1}{\sqrt {\calZ(\mu)}}\sum_{g,\bar{g}} e^{-2\mu|g+\bar{g}|}\kket{g,\bar g},
\end{align}
%
where $g+\bar{g}$ is a binary sum and represents the relative loop configuration, $\calZ(\mu)$ is the normalization factor, and $\mu$ can be interpreted as the loop tension.
%
We note that only the relative loop $h := g+\bar{g}$ acquires a loop tension.
%
Hence, it is more convenient to rewrite the EFD as
\begin{align}
\kket{\rho} = \frac{1}{\sqrt{\calZ(\mu)}} \sum_{g,h} e^{-2\mu|h|}\kket{g,h}.
\end{align}
The anyon condensation transition is detected by the open string operator
\begin{align}
\langle\!\langle W_{m\overline{m}}(\calP)\rangle\!\rangle = \frac{\bbra{\rho} w_{m}(\calP) \otimes w_{\overline{m}}(\calP) \kket{\rho}}{\bbrakket{\rho}{\rho}} = \frac{1}{\calZ(\mu)}\sum_{g,h} (-1)^{\# \text{crossing}(\calP,h)} e^{-4\mu|h|},
\end{align}
where $\# \text{crossing}(\calP,h)$ counts the number of crossings between the open string $\calP$ and loop $h$.
%
In this expression, the open string $\langle\!\langle W_{m\overline{m}}(\calP)\rangle\!\rangle$ can be interpreted as an observable in a statistical mechanical model of loops.
%
The normalization factor $\calZ(\mu)$ is the partition function $\calZ(\mu) = \Omega(0)\Omega(\mu)$, where the partition function $\Omega(0)$ of $g$ loops is a constant independent of $\mu$, and the partition function of $h$ loop, $\Omega(\mu) = \sum_h e^{-H(\mu)}$, has an effective Hamiltonian $H(\mu) = 4\mu|h|$.
We can identify the loop configuration $h$ with domain walls in a spin model with Ising degrees of freedom living on the plaquettes, i.e.
\begin{align}
\begin{tikzpicture}[scale=0.75, baseline={(current bounding box.center)}]
\definecolor{myred}{RGB}{240,83,90};
\definecolor{myblue}{RGB}{93,123,189};
\small
\foreach \y in {0,0.7,1.4,2.1}{
\draw[black!35] (-0.2,\y) -- (2.2,\y);
}
\foreach \x in {0,0.7,1.4,2.1}{
\draw[black!35] (\x,-0.2) -- (\x,2.2);
}
\draw[myblue,line width=0.6] (-0.2,0.7) -- (0.7,0.7) -- (0.7,1.4) -- (2.1,1.4) -- (2.1,2.2);
\draw[myblue,line width=1.5] (0.7,1.4) -- (1.4,1.4);
\filldraw (1.05,1.05) circle (0.03) node[right]{\scriptsize$\sigma_i$};
\filldraw (1.05,1.8) circle (0.03) node[right]{\scriptsize$\sigma_j$};
\draw[dotted,line width=1] (1.05,1.05) -- (1.05,1.8);
\node[myblue] at (0.3,1.8) {$h_{l}$};
\end{tikzpicture}, \qquad h_l = \frac{1 - \sigma_i\sigma_j}{2}.
\end{align}
%
Then, the loop model reduces to the square lattice Ising model up to a constant
\begin{align}
H = \sum_{\langle i, j\rangle} -2\mu \sigma_i\sigma_j.
\end{align}
Thus, the condensation transition in the EFD is of the 2D classical Ising universality class.
%
The critical point is exactly solvable and determined by the Kramers-Wannier duality, $\mu_c = (1/4)\log(1+\sqrt{2})$,
i.e. $p_c^{(2)} = 0.178$.
%
%
We remark that this critical point is associated with the partition function $\calZ(\mu) = \tr \rho^2$ and indicates the singularity in the second moment of $\rho$.
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
\subsection{Topological entanglement entropy}\label{sec:TEE}
Another diagnostic of the anyon condensation is the topological entanglement entropy (TEE) of the EFD state.
%
In the case of the Toric code subject to incoherent errors, the EFD state can be written as the ground state of a gapped local Hamiltonian.
%
One can therefore use TEE to detect the transition~\cite{}.
%
In this subsection, we first construct the parent Hamiltonian of the EFD.
%
We then compute the R\'enyi-$n$ entropy of the EFD and show that the subleading topological term undergoes a transition from $2\log 2$ to $\log 2$ upon the condensation of $m\bar{m}$ regardless of the R\'enyi order $n$.
%
This indicates a singularity also in the topological term of von Neumann entropy ($n\to 1$).
%
We remark that, in a related problem of the Toric code undergoing imaginary time evolution, the TEE changes sharply from $\log 2$ to zero across the phase transition as shown in Ref.~\cite{CastelnovoPRB2008}.
The EFD describing the Toric code subject to bit-flip errors is the ground state of a local parent Hamiltonian~\footnote{We thank Meng Cheng for pointing this out to us.}
\begin{align}
\tilde{H} = V^{-1} H_0 V V H_0 V^{-1},
\end{align}
where $V = e^{\mu \sum_i X_i \otimes X_i}$, and $H_0$ is a projector Hamiltonian given by
\begin{align}
H_0 = \sum_s \frac{1 - A_s\otimes \bbI}{2} + \frac{1 - \bbI \otimes A_s}{2} + \sum_p \frac{1 - B_p\otimes \bbI}{2} + \frac{1 - \bbI \otimes B_p}{2}.
\end{align}
%
%
Here, the eigenvalue of $\tilde{H}$ is non-negative. One can show $\kket{\rho}$ is a zero eigenstate of $\tilde{H}$ and therefore is the ground state.
%
It is then legitimate to use the TEE to characterize the topological order in the EFD.
%
The TEE sharply changes when the gap of $\tilde{H}$ closes.
We now compute the TEE in the EFD and show it undergoes a transition when tuning the error rate.
%
First, we show the TEE takes distinct values in two limits, the error rate $p = 0$ and $p = 1/2$.
%
To gain an understanding away from these two limits, we then consider the R\'enyi-$n$ entropy and map it to the free energy cost of spin pinning in a classical spin model.
%
We show its subleading term undergoes a transition at the critical point and take the analytic limit $n \to 1$ to extract the behavior of the TEE.
%
%
%
%
We remark that, as far as the entanglement entropy of a finite disk is concerned, the boundary condition of the total system is not important and will not be specified explicitly throughout the entire calculation.
In the limits $p = 0, 1/2$, the EFD state is a stabilizer state, and the TEE can be easily determined.
%
In the absence of errors, the EFD contains two copies of decoupled Toric code states and exhibits a TEE $\gamma = 2 \log 2$.
%
In the case of $p = 1/2$, the tension $\mu \to \infty$ imposes the stabilizers $X_i\otimes X_i$ in the EFD.
%
The resulting EFD is the mutual eigenstate of stabilizers
\begin{equation*}
\{A_s \otimes \mathbb{I}, \quad \mathbb{I} \otimes A_s, \quad X_i \otimes X_i, \quad B_p \otimes B_p\}.
\end{equation*}
For a simply connected region $A$, its entanglement entropy is given by
\begin{align}
S_A/\log 2 = 2|A| - \# \text{stabilizer}(\text{supp}(s) \in A)
\end{align}
where $2|A|$ counts the total number of degrees of freedom in region $A$, and $\# \text{stabilizer}(\text{supp}(s) \in A)$ is the number of stabilizers whose support is completely in $A$.
%
One can show $S_A = (|\partial A|-1)\log 2$, and the TEE $\gamma = \log 2$.
%
The distinct values of TEE in two limits indicate its potential singularity as a function of error rate.
To understand the TEE away from two limits, we consider the R\'enyi-$n$ entropies of a simply connected region $A$ and develop a classical spin model description.
%
The first step is to write down the reduced density matrix.
%
Here, we separate the loop configuration $g$ ($h$) into two parts $g_A$ and $g_{\bar{A}}$ ($h_A$ and $h_{\bar{A}}$), which support on region $A$ and its complement $\bar{A}$, respectively.
%
To form a closed loop configuration, two parts must agree on the boundary of $A$, i.e., $\partial g_A = \partial g_{\bar A} = b$, and $\partial h_A = \partial h_{\bar A} = c$.
\begin{equation*}
\begin{tikzpicture}[scale=0.45, baseline={(current bounding box.center)}]
\definecolor{myred}{RGB}{240,83,90};
\definecolor{myblue}{RGB}{93,123,189};
%
%
\definecolor{mygreen}{RGB}{83,195,189};
\small
\foreach \y in {0.7,1.4,...,4.9}{
\draw[black!40] (0.7-0.2,\y) -- (5.1,\y);
}
\foreach \x in {0.7,1.4,...,4.9}{
\draw[black!40] (\x,0.7-0.2) -- (\x,5.1);
}
%
\draw[myred, dashed, line width=1.5] (1.5,1.5) -- (4.1,1.5) -- (4.1,4.1) -- (1.5,4.1) -- cycle;
%
\draw[myblue, line width=1.5] (1.4,3.5) -- (2.1,3.5) -- (2.1,4.2);
\draw[myblue, line width=1.5] (2.1,1.4) -- (2.1,2.1) -- (2.8,2.1) -- (2.8,2.8) -- (3.5,2.8) -- (3.5,1.4);
%
\draw[mygreen, line width=1.5] (1.4,3.5) -- (0.7,3.5) -- (0.7,4.9) -- (2.1,4.9) -- (2.1,4.2);
\draw[mygreen, line width=1.5] (2.1,1.4) -- (2.1,0.7) -- (3.5,0.7) -- (3.5,1.4);
\node[myblue] at (2.9,3.3) {$g_A$};
\node[mygreen] at (1.4,0.9) {$g_{\bar A}$};
\end{tikzpicture}
\end{equation*}
%
%
%
%
%
%
%
The reduced density matrix can be written as
\begin{align}
\varrho_A(\mu) &= \Tr_{\bar{A}} \kket{\rho(\mu)}\bbra{\rho(\mu)} \nonumber \\
%
&= \frac{1}{\calZ(\mu)} \sum_{b, c} \sum_{\substack{g_A,\tilde{g}_A,\\\partial g_A = \partial \tilde{g}_A = b}} \sum_{\substack{
h_A, \tilde{h}_A, \\\partial h_A = \partial \tilde{h}_A = c}} \Omega_{\bar{A}}(0; b)\Omega_{\bar{A}}(\mu; c) e^{-2\mu|h_A| - 2\mu|\tilde{h}_A|} \kket{g_A, h_A}\bbra{\tilde{g}_A, \tilde{h}_A},
\end{align}
where $\Omega_{\bar{A}}(0;b)$ and $\Omega_{\bar{A}}(\mu;c)$ are the partition function of $g$ and $h$ loops in region $\bar{A}$ with a given boundary condition $b$ and $c$, respectively,
\begin{align}
\Omega_{\bar{A}}(\mu;c) = \sum_{h_{\bar{A}}, \, \partial h_{\bar{A}} = c} e^{-4\mu|h_{\bar{A}}|}.
\end{align}
%
The normalization factor $\calZ(\mu) = \Omega(0) \Omega(\mu)$ can be written accordingly as
\begin{align}
\Omega(\mu) = \sum_c \Omega_{A}(\mu;c) \Omega_{\bar{A}}(\mu;c).
%
\end{align}
Next, we write down the $n$-th moment of the reduced density matrix
\begin{align}
\Tr \varrho_A^n(\mu) = \Gamma_{A}(0) \Gamma_{A}(\mu),
\end{align}
where $\Gamma_{A}(0)$ and $\Gamma_A(\mu)$ are associated with $g$ and $h$ loops, respectively, and
\begin{align}
\Gamma_{A}(\mu) = \frac{1}{\Omega^n(\mu)}\sum_{c}\Omega_{A}^n(\mu;c) \Omega_{\bar{A}}^n(\mu;c).\label{eq:n_moment}
\end{align}
%
Equation~\eqref{eq:n_moment} can be viewed as a ratio between two partition functions.
%
The denominator represents the partition function of $n$ copies of loop models, while the loop model in the numerator is further subject to the constraint that loops in different copies have the same boundary condition $c$.
The R\'enyi entropy therefore can also be interpreted as the excess free energy related to the above constraint
\begin{align}
S_A^{(n)} = \frac{1}{1-n} \log \Tr\varrho_A^n(\mu) = \frac{1}{n-1}\left( F_A^{(n)}(0) - nF_0(0)\right) + \frac{1}{n-1}\left( F_A^{(n)}(\mu) - nF_0(\mu)\right)
\end{align}
where $F_0(\mu) = -\log\Omega(\mu)$ is the free energy of the loop model with tension $\mu$, $F_A^{(n)}(\mu) = -\log \sum_{c}\Omega_A^n(\mu;c)\Omega_{\bar{A}}^n(\mu;c)$ is the free energy of $n$ copies of the loop model subject to the constraint.
The scaling of the excess free energy is easier to analyze when formulating the loop model in terms of Ising spins.
%
We again identify the loop configuration with Ising domain walls
%
\begin{align}
g^{(s)}_l = \frac{1 - \sigma_i^{(s)}\sigma_j^{(s)}}{2}, \quad h_l^{(s)} = \frac{1 - \sigma'^{(s)}_i\sigma'^{(s)}_j}{2},\quad s = 1,2,\cdots, n
\end{align}
%
Furthermore, we redefine the Ising variables
\begin{align}
\sigma_i^{(s)} := \sigma_i^{(1)}\tau_i^{(s-1)}, \quad \sigma'^{(s)}_i := \sigma'^{(1)}_i\tau'^{(s-1)}_i,\quad s = 2, 3,\cdots, n.
\end{align}
%
The constraints on the loop configurations require $n$ copies of the Ising spins, $\sigma^{(s)}$ and $\sigma'^{(s)}$, on the boundary of the region $A$ having the same domain wall configurations.
%
Thus, they require the relative Ising spin $\tau^{(s)}_i$ and $\tau'^{(s)}_i$ having no domain wall on the boundary of $A$, i.e. pointing in the same direction.
%
There are $(n-1)$ such constraints in $F_A^{(n)}(\mu)$, therefore $F_A^{(n)}(\mu) = nF_0(\mu) + (n-1)\Delta F(\mu)$, where $\Delta F$ is the excess free energy for aligning one copy of the $\tau$ spins on the boundary of $A$.
%
Correspondingly, the R\'enyi entropy $S_A^{(n)} = \Delta F(0) + \Delta F(\mu)$.
The excess free energy has two contributions.
%
The energetic part is always proportional to $|\partial A|$.
%
The entropic part counts the loss of degrees of freedom due to spin aligning.
%
In the ferromagnetic phase, the spins can fluctuate below the scale of correlation length $\xi$.
%
Thus, $\Delta F \propto |\partial A|/\xi$ exhibits a strict area-law scaling.
%
In the paramagnetic phase, the spins can both fluctuate freely above the scale of correlation length $\xi$.
%
At the leading order, the free energy cost is proportional to $|\partial A|$.
%
However, the aligned spin can fluctuate globally giving rise to a subleading correction, i.e. $\Delta F = a|\partial A| - \log 2$.
%
Thus, the subleading term sharply changes at the ferromagnetic transition.
The above analysis indicates the R\'enyi entropy exhibits an area-law scaling $S_A^{(n)} = a |\partial A| - \gamma^{(n)}$, and the topological term $\gamma^{(n)}$ distinguishes two phases of the effective spin model.
%
When $\mu < \mu_c$, both spin models for $\sigma$ and $\sigma'$ are in the paramagnetic phase, giving rise to $\gamma^{(n)} = 2\log 2$.
%
When $\mu > \mu_c$, the spin model for $\sigma'$ (i.e. the loop model for $h$) is in the ferrmagnetic phase, reducing $\gamma^{(n)}$ to $\log 2$.
%
The result holds for an arbitrary R\'enyi order $n$ and also in the replica limit $n \to 1$.
%
Hence, the TEE of the EFD state sharply changes from $2\log 2$ to $\log 2$ when $m\bar{m}$ condenses at $p_c^{(2)}$.
We make an important remark that the effective model for the R\'enyi-$n$ entropy of the EFD is merely $n$ decoupled loop models.
%
The partition function $\calZ^n(\mu) = (\tr\rho^2)^n$.
%
Therefore, the transition occurs at the same critical threshold $p_c^{(2)}$ regardless of $n$.
%
This is fundamentally different from the transition in the $n$-th moment $\tr \rho^n$ of the corrupted density matrix, which occurs at a different critical threshold $p_c^{(n)}$.
The Toric code state subject to Pauli-Y and Z errors can be similarly analyzed.
%
These errors can induce a condensation of $f\bar{f}$ and $e\bar{e}$ in the EFD at the same critical threshold $p_c^{(2)}$.
%
We also note that, anyon bound states $\alpha\bar{\alpha}$ and $\beta\bar{\beta}$ are mutual bosons and therefore can condense independently.
%
When two types of anyons are condensed, e.g. $e\bar{e}$ and $m\bar{m}$, the rest one $f\bar{f}$ is also condensed as $e \times m = f$.
%
The resulting EFD no longer contains condensed loop objects and will exhibit a vanishing TEE.
\section{Mapping decoherence-induced phases to boundary anyon condensates}
In this section, we map the decoherence-induced phases in the EFD to the boundary phases of a topologically ordered system.
%
In Sec.~\ref{sec:path_integral}, we start with the path integral formulation of the EFD state.
%
In Sec.~\ref{sec:mapping}, we map the decoherence-induced phases to the boundary phases of topologically ordered systems.
%
In particular, we establish the classification scheme for the possible phases in abelian topological order subject to incoherent errors.
%
In Sec.~\ref{sec:examples}, we carry out the classification in three examples, the Toric code, double semion, and $\nu = 1/3$ Laughlin state with incoherent errors.
%
The results are summarized in Table~I of the main text.
%
Section~\ref{sec:nth} generalizes the field theory description for the second moment to the $n$-th moment $\tr\rho^n$.
%
We remark that, for abelian topological order with incoherent errors, the number of possible phases is independent of $n$.
%
The possible phases in the EFD have one-to-one correspondence to that in the replica limit $n\to 1$.
\subsection{Path integral formulation of the EFD}\label{sec:path_integral}
Here, we develop a path integral formulation of the EFD $\kket{\rho}$.
%
The double ground state $\ket{\Psi_0}\otimes\ket{\Psi_0^*}$ in the EFD can be prepared using an infinite imaginary time evolution from an arbitrary state, which is equivalently a Euclidean path integral in spacetime region $\tau < 0$.
%
The wavefunction component at $\tau = 0$ can be formally written as
%
\begin{equation}
\begin{aligned}
\calZ_{\tau<0} &[\varphi, \bar{\varphi}] := \big( \bra{\varphi} \otimes \bra{\bar{\varphi}} \big) \big(\ket{\Psi_0} \otimes\ket{\Psi_0^*} \big) = \int\displaylimits_{(\varphi',\bar\varphi')|_{\tau = 0^-} = (\varphi,\bar\varphi)} \calD (\varphi',\bar{\varphi}') e^{-\calS(\varphi') - \calS^*(\bar\varphi')}
\end{aligned},
\end{equation}
where $\varphi$ and $\bar\varphi$ are the dynamical variables in the ket and bra Hilbert space, the path integral is defined in the past spacetime region $\tau<0$, $\calS(\varphi')$ and $\calS^*(\bar\varphi')$ are a pair of conjugate Euclidean actions that yield the ground states $\ket{\Psi_0}$ and $\ket{\Psi_0^*}$, respectively.
%
The EFD $\kket{\rho}$ is then obtained by turning on the error channel which couples the field $\varphi$ and $\bar\varphi$ at $\tau=0$ time slice locally [illustrated in Fig.~1(b) of the main text].
%
%
The norm $\bbrakket{\rho}{\rho}$ is accordingly written as a Euclidean path integral in the entire spacetime, i.e. two path integrals in the future ($\tau>0$) and past ($\tau<0$) glued at $\tau = 0$:
%
\begin{equation}
\bbrakket{\rho}{\rho} = \int \calD(\varphi_p, \bar{\varphi}_p, \varphi_f, \bar{\varphi}_f) \, \calZ_{\tau>0}[\varphi_f, \bar\varphi_f] \calZ_{\tau<0}[\varphi_p, \bar\varphi_p] e^{- \calS_{int}(\varphi_f,\bar\varphi_f,\varphi_p,\bar\varphi_p)},
\label{eq:path_int}
\end{equation}
where $\varphi_{f(p)}$ and $\bar{\varphi}_{f(p)}$ denote the future (past) field variables at the time slice $\tau=0^+$ ($\tau = 0^-$).
%
The future and past fields are coupled by the error channel, and the coupling is formally written as
$$
\begin{aligned}
e^{-\calS_{int}(\varphi_f,\bar\varphi_f,\varphi_p,\bar\varphi_p)} = \bbra{\varphi_f, \bar\varphi_f }\calN^\dag \calN \kket{ \varphi_p, \bar\varphi_p }\,.
\end{aligned}
$$
In the absence of errors, $\calN = \bbI$ and $e^{-\calS_{int}} = \delta_{\varphi_p,\varphi_f} \delta_{\bar{\varphi}_p,\bar{\varphi}_f}$ becomes a constraint.
%
We further remark that individual Kraus operators and $\calN$ are generally non-Hermitian.
%
However, $\calS_{int}$ only contains terms determined by the Hermitian combination $\calN^\dag \calN$, which guarantees $\bbrakket{\rho}{\rho}$ being a real number.
\subsection{Mapping to (1+1)D boundary phases}\label{sec:mapping}
In this subsection, we take advantage of the path-integral formulation and work with the topological quantum field theory (TQFT) describing the long-distance physics of topological orders.
%
The invariance of TQFT under the $\pi/2$ spacetime rotation allows mapping decoherence-induced phases in the EFD to $(1+1)$D quantum phases on the edge of topologically ordered systems.
%
Here, we consider specifically the abelian topological order described by the Chern-Simons (CS) theory.
%
In the case of incoherent errors, we carry out a classification of the possible decoherence-induced phases using the effective theory on the edge.
%
We demonstrate our general ideas with two concrete examples, the Toric code and double semion, and find five phases in each model.
%
The possible phases in the case of coherent errors are also commented.
%
%
%
%
%
%
%
%
%
%
%
%
%
We consider the EFD for the Abelian topologically ordered ground state subject to incoherent errors.
%
The ground state $\ket{\Psi_0}$ is described by a $U(1)^M$ Chern-Simons theory with the following Lagrangian
\begin{equation}
\calL_{CS}[a] = \frac{1}{4\pi} \sum_{I,J = 1}^{M} K_{IJ}\, a^I \wedge \rd a^J + \cdots \,,
\end{equation}
where $K$ is a symmetric $M \times M$ integer matrix and $a^{I=1,2,\ldots,M}$ are the $U(1)$ gauge fields, the ellipsis denotes parts that involve matter fields~\cite{Read:1990zz,Wen:1992uk,Frohlich:1991wb,Wen:1995qn}.
%
Matter fields are gapped and do not enter the dynamics at long distances, yet, they are necessary in giving rise to the anyonic excitations.
%
Accordingly, $\ket{\Psi_0^*}$ carries the conjugate topological order and is described by the same Lagrangian except that the $K$-matrix has an additional minus sign. To better distinguish them, we use $\bar{a}^{I=1,2,\ldots,M}$ to denote the gauge fields for $\ket{\Psi_0^*}$.
%
We can combine these two parts and write the path integral $\calZ_{\tau<0}$ or $\calZ_{\tau>0}$ in terms of the following Lagrangian
\begin{equation}
\calL_{CS}[\mathbf{a}] = \frac{1}{4\pi} \sum_{I,J = 1}^{2M} \mathbb{K}_{IJ}\, \mathbf{a}^I \wedge \rd \mathbf{a}^J + \cdots,
\end{equation}
where $\mathbb{K} = K \oplus (-K)$, and $\mathbf{a} = [a, \bar{a}]^T$ is a $2M$ component vector of the gauge fields in the Hilbert space $\calH\otimes \calH$.
%
%
%
%
Studying the decoherence-induced phases and the potential phase transitions in the EFD in the current form is not convenient.
%
On one hand, in general, anyonic excitations created by local errors explicitly involve matter fields and are complicated to analyze.
%
On the other hand, the transition, which happens at the temporal interface, is different from a ground state problem and largely unexplored.
These technical difficulties necessitate our following trick.
%
We perform a spacetime rotation to exchange the imaginary time $\tau$ and the spatial coordinate $x$ [Fig.~1(c) of the main text], i.e.,
\begin{equation}
\tau \rightarrow -x\,,\quad
x \rightarrow \tau\,,\quad
y\rightarrow y\,,
\end{equation}
where $x,y$ denote the spatial coordinates.
%
The rotation turns the temporal interface at $\tau = 0$ into a spatial one at $x = 0$.
%
The original path integral \eqref{eq:path_int} can be formally rewritten as
\begin{equation*}
\bbrakket{\rho}{\rho} = \int \calD(\varphi_L, \bar{\varphi}_L, \varphi_R, \bar{\varphi}_R) \, \calZ_{x>0}[\varphi_R, \bar\varphi_R] \calZ_{x<0}[\varphi_L, \bar\varphi_L] e^{- \tilde \calS_{int}(\varphi_R,\bar\varphi_R,\varphi_L,\bar\varphi_L)}\,,
\end{equation*}
where $\calZ_{x<0}$, obtained by rotating $\calZ_{\tau<0}$, is the path integral on the left side of the spatial interface $x = 0^-$, and $(\varphi_L,\bar{\varphi}_L)$ specify the value of the fields at the spatial interface. The meaning of $\calZ_{x>0}$ and $(\varphi_R,\bar{\varphi}_R)$ is similar.
$\tilde \calS_{int}$ describes the coupling between the two half systems at the spatial interface and generally takes a different form from the original one $\calS_{int}$.
After the rotation, the path integral is translationally invariant in the new time direction and describes the norm of a $(2+1)$D ground state [Fig.~1(c) of the main text].
%
In the rotated picture, the decoherence-induced phases correspond to the $(1+1)$D phases on the 1D defect at $x = 0$.
%
%
%
%
%
%
The Lagrangian that describes the 1D defect has three contributions
\begin{align}
\calL = \calL_0 + \calL_1 + \calL_{\calN},
\end{align}
where $\calL_0$ describes the low energy edge excitations for the $(2+1)D$ theory in the region $x > 0$ and $x < 0$.
%
The interaction on the defect is derived from $\tilde{S}_{int}$ and contains two parts.
%
The first part originates from the definition of path integral and couples degrees of freedom on the left and the right given by $\calL_1$, which is present even in the absence of a quantum channel.
%
The second part $\calL_\calN$ is introduced by the quantum channel.
%
In what follows, we explicitly write down the three terms in the Lagrangian and analyze the possible phases.
We start with the low energy edge theory $\calL_0$ for the CS theory in the bulk.
%
We consider how the path integral for the past $\calZ_{\tau<0}$ changes under the rotation:
\begin{itemize}
\item Originally, $\calZ_{\tau<0}[\varphi_p, \bar{\varphi}_p]$ describes the path integral that prepares the ground state of a system on a whole plane, where $\varphi_p, \bar{\varphi}_p$ specifies the component of the wavefunction. Note that $\varphi_p$ and $\bar{\varphi}_p$ includes both the CS gauge fields and the matter fields, and thus they are complicated to deal with.
\item The spacetime rotation transforms it into $\calZ_{x>0}[\varphi_R, \bar{\varphi}_R]$, a path integral defined on the half-space but the entire time domain, and $\varphi_R, \bar{\varphi}_R$ specifies the boundary value of the fields.
%
The bulk part of the path integral contains the same CS action as that before the rotation and the part involving matter fields.
%
The boundary has a simple description in terms of compact bosons, which is the main reason why the rotation trick can be useful.
\item From now on, we will replace $\varphi_R$ and $\bar{\varphi}_R$ by the compact boson fields $\phi_R, \bar{\phi}_R$, which correspond to the bulk gauge fields $a, \bar{a}$.
Then, the low-energy dynamics of the boundary of the CS theory is governed by the following effective Lagrangian
\begin{equation}
\calL_R[\Phi_R] = \frac{1}{4\pi} \sum_{I,J=1}^{2M} -\mathbb{K}_{IJ} \ri \partial_\tau \Phi_R^I \partial_y \Phi_R^J - \mathbb{V}_{IJ} \partial_y \Phi_R^I \partial_y \Phi_R^J,
\end{equation}
where $\Phi_R := [\phi_R, \bar{\phi}_R]$, $\mathbb{K}$ is the same K-matrix as the one for the bulk CS action, and $\mathbb{V}$ is a positive semi-definition matrix that determines the velocity of the boson fields.
The path integral with fixed boundary values of the boson fields is then determined by $\calL_R[\Phi_R]$ classically:
\begin{equation}
\calZ_{x<0}[\Phi_R] = \exp\left( - \int \rd\tau \rd y \calL_R[\Phi_R] \right)\,.
\end{equation}
\end{itemize}
The transformation of the path integral for the future $\calZ_{\tau>0}$ can be analyzed similarly. The result is related to what we have obtained above by a spatial reflection
\begin{equation*}
\calL_L[\Phi_L] = \frac{1}{4\pi} \sum_{I,J=1}^{2M} \mathbb{K}_{IJ} \ri \partial_\tau \Phi_L^I \partial_y \Phi_L^J - \mathbb{V}_{IJ} \partial_y \Phi_L^I \partial_y \Phi_L^J,
\end{equation*}
where $\Phi_L := [\phi_L, \bar{\phi}_L]$, and the minus sign in front of the $K$-matrix comes from the spatial reflection.
The path integral $\calZ_{x>0}[\Phi_L]$ can also be determined classically
\begin{equation}
\calZ_{x>0}[\Phi_L] = \exp\left( - \int \rd\tau \rd y \calL_L[\Phi_L] \right)\,.
\end{equation}
%
%
%
We combine the low-energy excitations from the left and the right and obtain the Lagrangian describing the edge of the quadruple topological order
\begin{equation}
\mathcal{L}_0[\phi] = \frac{1}{4\pi} \sum_{I,J} \mathbb{K}_{IJ}^{(2)} \ri \partial_\tau \phi^I \partial_y \phi^J - \mathbb{V}_{IJ}^{(2)} \partial_y \phi^I \partial_y \phi^J\,,\label{eq:edge_lagrangian}
\end{equation}
where the $4M\times 4M$ $K$-matrix $\mathbb{K}^{(2)} = \mathbb{K} \oplus \mathbb{K}$, velocity matrix $\mathbb{V}^{(2)} = \bbV\oplus\bbV$, and $\phi := [\bar{\phi}_R, \phi_R, \phi_L, \bar{\phi}_L]$.
%
%
%
%
%
%
%
Next, we discuss the interaction in the effective theory.
%
The first part $\calL_1$ is associated with the definition of path integral and couples $\phi_L$ to $\phi_R$ and $\bar{\phi}_L$ to $\bar{\phi}_R$.
%
It leads to a gapped phase of the defect in the absence of errors.
%
The second part $\calL_\calN$ is introduced by the error channel.
%
It couples the bosonic fields from the same copy of the density matrix (or equivalently EFD), namely $\phi_s$ to $\bar{\phi}_s$ for $s = L, R$.
To explicitly write down the interaction in terms of the compact bosons, we first list the excitations in the effective theory $\calL_0$.
%
The excitations are labeled by integer vectors $\bm{l}_\alpha$ and are created by $e^{\ri\bm{l}_\alpha^T\cdot \phi}$.
%
They have anyonic statistics, and the braiding of two anyons $\bm{l}_\alpha$ and $\bm{l}_\beta$ yields a phase
\begin{equation}
\theta_{\alpha\beta} = 2\pi \bm{l}_\alpha^T (\mathbb{K}^{(2)})^{-1} \bm{l}_\beta.
\end{equation}
%
We remark that, in general, a single anyon cannot be created using a local operator.
%
Instead, a pair of anyon $\alpha$ and its antiparticle $\alpha'$ can be created using a local operator, which is described by $\Psi_{\Lambda} := e^{\ri (\mathbb{K}^{(2)} \Lambda)^T \cdot \phi}$ with $\Lambda$ being an integer vector.
%
Such excitation has trivial braiding with other excitations, i.e. $\theta_{\alpha, \mathbb{K}^{(2)}\Lambda} = 0 \mod 2\pi, \forall \alpha$.
%
%
%
%
Furthermore, the form of the interaction is limited by the global $\calG^{(2)} = \mathbb{Z}_2 \times \mathbb{Z}_2^\mathbb{H}$ symmetry of 1D defect.
%
The first $\mathbb{Z}_2$ is an anti-unitary symmetry generated by the hermitian conjugation in the double Hilbert space.
%
The path integral represents the norm of the EFD, which is a real number.
%
Thus, the effective theory is invariant under the transformation
%
\begin{equation}
\begin{aligned}
\phi_{L}^I &\leftrightarrow \phi_{R}^I, \quad \bar{\phi}_{L}^I \leftrightarrow \bar{\phi}_{R}^I.
\end{aligned}
\end{equation}
%
In addition, the EFD represents the density matrix, which is hermitian in the original Hilbert space.
%
Thus, edge theory exhibits another anti-unitary $\mathbb{Z}_2^\mathbb{H}$ symmetry due to the hermiticity of the density matrix, which acts as
\begin{equation}
\begin{aligned}
\phi_{s}^I &\leftrightarrow \bar{\phi}_{s}^I,\quad s = L,R.
\end{aligned}
\end{equation}
We remark that the K-matrix $\mathbb{K}^{(2)}$ flips the sign under the anti-unitary symmetry, however, the Lagrangian $\calL_0$ is invariant as the minus sign cancels that from the imaginary identity.
We are now ready to explicitly write down the interaction in terms of edge excitations.
%
The first part $\calL_1$ from the definition of path integral can be written in terms of the back-scattering between the left and the right fields.
%
It is generally given by a hopping term between the left and the right
\begin{align}
\sum_{\Lambda} \frac{B_{\Lambda}(y)e^{\ri \theta(y)}}{2} \left(\Psi_{L,\Lambda}^\dagger \Psi_{R,\Lambda} + \bar{\Psi}_{R,\Lambda}^\dagger \bar{\Psi}_{L,\Lambda}\right) + h.c.,
\end{align}
where $\Psi_{L,\Lambda} := e^{\ri (K\Lambda)^T \phi_L}$, $\Psi_{R,\Lambda} := e^{-\ri (K\Lambda)^T \phi_R}$, $\bar{\Psi}_{L,\Lambda} := e^{\ri (-K\Lambda)^T \bar{\phi}_L}$, and $\bar{\Psi}_{R,\Lambda} := e^{\ri (K\Lambda)^T \bar{\phi}_R}$ are local operators on the edge, $\Lambda$ is a $M$-component integer vector, the real coefficient $B_\Lambda$ and phase factor $\theta$ have spatial dependence in general.
%
%
%
%
%
In terms of the compact boson fields, the back-scattering takes the form
\begin{equation}
\mathcal{L}_1 = \sum_\Lambda B_\Lambda \Big[ \cos\Big(\sum_{I = 1}^M (K\Lambda)_I (\phi_L^I + \phi_R^I) +\theta \Big) + \cos\Big(\sum_{I = 1}^M (K\Lambda)_{I}(\bar{\phi}_L^I + \bar{\phi}_R^I)+\theta\Big)\Big].\label{eq:edge_int_ref}
\end{equation}
%
We remark that the coefficients for two cosine terms are forced to the same real number due to the global symmetry $\mathcal{G}^{(2)}$.
%
With large coefficients, such cosine terms tend to pin the bosonic fields $\phi+\bar{\phi}$ and lead to a gapped phase of the defect without any error channel.
%
These interaction terms only create anyon bound states on the left and the right in the form $\alpha_L\alpha'_R = e^{\ri\bm{l}^T\cdot(\phi_L + \phi_R)}$ and $\bar{\alpha}_L\bar{\alpha}'_R = e^{-\ri\bm{l}^T\cdot(\bar{\phi}_L + \bar{\phi}_R)}$.
The second part $\calL_\calN$ of the interaction is introduced by the quantum channel and couples the fields from the same copy of the density matrix.
%
%
%
%
%
For incoherent error channels, before the rotation, only anyon bound states $\alpha\bar{\alpha}$ are created in the EFD.
%
In the rotated picture, the quantum channel creates excitations $\alpha_s\bar{\alpha}_s = e^{\ri\bm{l}^T \cdot(\phi_s-\bar{\phi}_s)}$.
%
At the 1D defect, it corresponds to the cosine term
\begin{align}
\mathcal{L}_{\mathcal{N}} = \sum_\Lambda C_\Lambda \sum_{s = L, R} \cos\left( \sum_{I = 1}^M (K\Lambda)_I (\phi_{s}^I - \bar{\phi}_{s}^I) \right),\label{eq:int_quantum_channel}
\end{align}
where $C_\Lambda$ is a real coefficient with spatial dependence.
%
%
%
We remark that the coefficient of the different cosine terms for $s = L, R$ are the same due to the $\mathbb{Z}_2$ symmetry.
%
%
%
%
%
%
The possible phases of the defect are classified by the inequivalent ways of gapping out $\calL_0$ in Eq.~\eqref{eq:edge_lagrangian} by condensing bosonic excitations.
%
For abelian topological order, the bosonic objects form a group, dubbed the Lagrangian subgroup $\calM$~\cite{Levin:2013gaa,Barkeshli:2013}.
%
Thus, the phases of abelian topological order subject to incoherent errors are classfied by the subgroup $\calM$ that satisfying the following criteria:
\begin{enumerate}
\item $e^{\ri \theta_{\bm{m}\bm{m'}}} = 1, \forall \bm{m}, \bm{m'} \in \calM$;
\item $\forall \bm{l} \notin \calM$, $\exists \bm{m}$ s.t. $e^{\ri \theta_{\bm{ml}}} \neq 1$;
\item $\forall \bm{m} \in \calM$, $ g\bm{m} g^{-1} \in \calM, \forall g \in \calG^{(2)}$;
\item (Incoherent error) $[\bm{1}, \bm{1},-\bm{1},-\bm{1}]^T \cdot \bm{m} = 0 \mod \mathbb{K}^{(2)}\Lambda, \forall \bm{m} \in \calM$,
\end{enumerate}
where $\Lambda$ is a $4M$-component integer vector, $\bm{1}$ is an $M$-component vector with each element being unity.
%
Here, the third criterion originates from the symmetry constraints, i.e. the excitations related by symmetry transformations must condense simultaneously.
%
The last criterion is due to incoherent errors.
%
The interaction on the edge only creates anyon bound states $\alpha_L\bar{\alpha}_L$, $\alpha_R\bar{\alpha}_R$, $\alpha_L\alpha'_R$, $\bar{\alpha}_L\bar{\alpha}'_R$ and their fusion results.
%
Therefore, the condensed objects satisfy the last criterion.
In the case of coherent errors, the phases are still classified by the Lagrangian subgroup.
%
However, since the Kraus operators can, for example, create anyon excitation $\alpha$ and $\bar{\alpha}$ in a single copy, the condensed objects need not to satisfy the last criterion.
%
%
%
Correspondingly, in the Lagrangian, the coherent errors can introduce the cosine terms, $\cos((K\Lambda)^T \cdot \phi_{s})$ and $\cos((K\Lambda)^T \cdot \bar{\phi}_{s})$, which solely depends on the compact boson in an individual copy of the topological order.
%
This can possibly lead to additional decoherence-induced phases.
Each edge condensate corresponds to a distinct phase of the EFD, and its condensed objects also determine the capability of encoding information in the error-corrupted state.
%
In the main text, we mention that the logical operator $W_{\alpha\bar{\alpha}}(\ell)$ can no longer encode quantum information in the condensate of anyon bound state $\alpha\bar{\alpha}$ in the EFD, which is equivalent to condensing $\alpha_s\bar{\alpha}_s$ with $s = L,R$ in the $(1+1)$D theory of the defect.
%
Hence, examining whether $\alpha_s\bar{\alpha}_s$ is condensed on the edge can determine the encoding properties in the error-corrupted state.
\subsection{Examples}\label{sec:examples}
Here, we classify the possible phases by enumerating the Lagrangian subgroups in three examples, 2D Toric code, double semion model, and $\nu = 1/3$ Laughlin state, subject to incoherent errors.
%
We comment on the information encoding in each phase.
%
The results are summarized in Table~I of the main text.
\emph{Example 1: Toric code.}---
The first example we consider is the 2D Toric code subject to incoherent errors.
%
We find five distinct gapped phases [see Table~I~\footnote{Here and in the following, Table~I refers to the table in the main text.}].
To start with, the low energy excitations in 2D Toric code are described by the CS theory with K-matrix
\begin{align}
K_{TC} = \begin{pmatrix}
0 & 2 \\
2 & 0
\end{pmatrix}.
\end{align}
The Toric code has four superselection sectors: $1, e, m, f$. Here, $e$ and $m$ are self-boson and exhibit mutual semionic statistics. The fermion $f$ is a composite object of $e$ and $m$.
%
These anyonic quasiparticles are labeled by integer vectors
\begin{align}
\bm{l}_e = \begin{pmatrix}
1 \\
0
\end{pmatrix}, \quad \bm{l}_m = \begin{pmatrix}
0 \\
1
\end{pmatrix}, \quad \bm{l}_f = \begin{pmatrix}
1 \\
1
\end{pmatrix}.
\end{align}
The first gapped phase is achieved when no error is present.
%
In this case, the definition of path integral introduces the cosine terms in Eq.~\eqref{eq:edge_int_ref}
\begin{equation}
\begin{aligned}
\calL_{1} = \sum_{I=1,2} B_I \Big[&\cos\left(2(\phi_L^I + \phi_R^I ) + \theta\right) + (\phi_s \leftrightarrow \bar{\phi}_s) \Big] + \cdots.
\end{aligned}\label{eq:toric_code_int_ref}
\end{equation}
%
%
Here, the ellipses represent the less relevant terms.
%
With sufficiently large coefficients $B_I$, these cosine terms can induce an edge condensate of $e_{L}e_R, \bar{e}_L\bar{e}_R, m_Lm_R$, and $\bar{m}_L\bar{m}_R$ (Phase I in Table~I).
%
The corrupted density matrix in this phase can encode quantum information because $\alpha_s\bar{\alpha}_s$ is not condensed.
The incoherent error channel can induce distinct edge phases.
%
%
Here, we have three cosine terms in the form of Eq.~\eqref{eq:int_quantum_channel}
\begin{equation}
\begin{aligned}
\mathcal{L}_{\mathcal{N},e} &= C_e\sum_{s = L, R} \cos\left(2(\bm{l}_e^T \cdot \phi_{s}-\bm{l}_e^T \cdot \bar{\phi}_{s})\right),\\
\mathcal{L}_{\mathcal{N},m} &= C_m\sum_{s = L, R} \cos\left(2(\bm{l}_m^T \cdot \phi_{s}-\bm{l}_m^T \cdot \bar{\phi}_{s})\right),\\
\mathcal{L}_{\mathcal{N},f} &= C_f\sum_{s = L, R} \cos\left(2(\bm{l}_f^T\cdot \phi_{s}-\bm{l}_f^T\cdot \bar{\phi}_{s})\right),
\end{aligned}\label{eq:toric_code_int_error}
\end{equation}
which are generated by the incoherent channels that create $e$, $m$, and $f$, respectively.
%
With a sufficiently large coefficient, any one of three cosine terms can develop a classical value leading to a condensate of $\alpha_s \bar{\alpha}_s$ with $\alpha = e, m, f$, which corresponds to Phase II-IV in Table~I, respectively.
%
In these phases, the remaining logical operators mutually commute, and the corrupted state is a classical memory.
Finally, when two cosine terms develop a classical value, the remaining one also achieves a classical value (Phase V in Table~I).
%
Here, the corrupted state is a trivial state, and one cannot encode any information.
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
Allowing coherent errors may lead to additional gapped phases beyond the above five possibilities.
%
For example, the amplitude-damping channel
\begin{align}
\calK_{1,i} = \frac{1+\sqrt{1-\kappa}}{2}\bbI + \frac{1-\sqrt{1-\kappa}}{2} Z_i, \quad \calK_{2,i} = \frac{\sqrt{\kappa}}{2}(X_i+\ri Y_i),
\end{align}
which can generate $e_s, \bar{e}_s, m_s, \bar{m}_s, e_s\bar{m}_s, m_s\bar{e}_s$ for $s = L, R$ in the effective theory.
%
These anyons cannot be created by incoherent errors;
condensing a subset of them can lead to distinct phases.
%
This possibly leads to additional phases of the 1D defect labeled by the Lagrangian subgroups with following generators:
\begin{itemize}
\item $e_s, \bar{e}_s \text{ for } s = L, R$;
\item $m_s, \bar{m}_s \text{ for } s = L, R$;
\item $e_s\bar{m}_s, m_s\bar{e}_s \text{ for } s = L, R$.
\end{itemize}
%
%
%
However, as the coherent channel creates all these anyons together with the anyon bound state $\alpha_s\bar{\alpha}_s$, these possible phases will compete.
%
It requires numerical simulation to determine which anyon condenses at low energy.
%
It remains open whether a specific coherent error channel can realize any of these phases.
We make a few remarks.
%
First, the phase with $e$ or $m$ anyon condensed in individual copies can be obtained by an imaginary time evolution of the Toric code ground state.
%
For example, $\ket{\Psi_0} \rightarrow e^{\mu \sum_i Z_i}\ket{\Psi_0}$ can realize the edge condensate of $e_s$ and $\bar{e}_s$.
%
Second, the above three phases are studied in the context of 1D defect phases in double Toric code and are referred to as the ``cut-open" and ``SPT" boundary~\cite{Lichtman:2020nuw}.
\emph{Example 2: Double semion.}---
The second example we consider is the double semion model subject to incoherent errors~\cite{LevinGu:2012}.
%
Again, we enumerate the possible Lagrangian subgroups that characterize the distinct phases.
%
We find five phases summarized in Table~I.
The low-energy theory for double semion topological order is the abelian CS theory with K-matrix
\begin{align}
K_{DS} = \begin{pmatrix}
2 & 0 \\
0 & -2
\end{pmatrix}.
\end{align}
The double semion model has four superselection sectors: $1, m_a, m_b, b$, where $m_a$ is a semion, $m_b$ is an anti-semion, and the boson $b$ is a composite object of $m_a$ and $m_b$.
%
The semion $m_a$ and anti-semion $m_b$ are mutual boson.
%
These anyons are characterized by the integer vectors
\begin{align}
\bm{l}_{m_a} = \begin{pmatrix}
1 \\
0
\end{pmatrix}, \quad \bm{l}_{m_b} = \begin{pmatrix}
0 \\
1
\end{pmatrix}, \quad \bm{l}_b = \begin{pmatrix}
1 \\
1
\end{pmatrix}.
\end{align}
The double semion model on the torus has a four-fold degenerate ground state that can encode two logical qubits.
%
The logical operators are the loop operators $w_\alpha(\ell)$ transporting $\alpha = m_a, m_b$ along two inequivalent large loops $\ell = \ell_1,\ell_2$.
%
Distinct from the case of Toric code, the operators along different loops, $w_\alpha(\ell_1)$ and $w_\alpha(\ell_2)$ for $\alpha = m_a, m_b$, anti-commute due to the anyon self-statistics and therefore serve as Pauli-X and Z operators for a logical qubit.
The first gapped phase (Phase I in Table~I) of the defect is realized when no error is present.
%
The backscattering term $\calL_1$ condenses the anyon bound states $\alpha_L{\alpha}'_R$ and $\bar{\alpha}_L\bar{\alpha}'_R$ for $\alpha = m_a, m_b, b$.
%
This phase can encode two qubits of quantum information.
The incoherent channel in the double semion model can introduce the following cosine terms:
\begin{equation}
\label{eq:double_semion_int_error}
\begin{aligned}
\mathcal{L}_{\mathcal{N},m_a} &= C_{m_a}\sum_{s = L, R} \cos\left(2(\bm{l}_{m_a}^T\cdot\phi_{s}-\bm{l}_{m_a}^T\cdot\bar{\phi}_{s})\right), \\
\mathcal{L}_{\mathcal{N},m_b} &= C_{m_b}\sum_{s = L, R} \cos\left(2(\bm{l}_{m_b}^T\cdot\phi_{s}-\bm{l}_{m_a}^T\cdot\bar{\phi}_{s})\right),\\
\mathcal{L}_{\mathcal{N},b} &= C_b\sum_{s = L, R} \cos\left(2(\bm{l}_{b}^T\cdot\phi_{s}-\bm{l}_{b}^T\cdot\bar{\phi}_{s})\right),
\end{aligned}
\end{equation}
which creates anyon bound states $\alpha_s\bar{\alpha}_s$ for $\alpha = m_a, m_b, b$, respectively.
%
Condensing each one of these terms induces Phase II-IV in Table~I, respectively.
%
In Phase II [III], one can encode information in $w_{m_a}(\ell)$ [$w_{m_b}(\ell)$] for $\ell = \ell_1,\ell_2$.
%
These logical operators do not mutually commute and therefore can encode one qubit of quantum information.
%
In Phase IV, the corrupted state can still encode one qubit of quantum information, despite that $w_{m_a}$ and $w_{m_b}$ are identified due to the condensation of $b_s\bar{b}_s$.
%
Condensing two terms in Eq.~\eqref{eq:double_semion_int_error} indicates the third is condensed and leads to Phase V, which cannot encode information.
Introducing coherent errors can potentially realize an additional phase by condensing the boson in individual copies, $b_s$ and $\bar{b}_s$.
%
Again, designing a specific coherent error channel to possibly realize this phase is left for future study.
\emph{Example 3: $\nu = 1/3$ Laughlin state.---}
The third example is $\nu = 1/3$ Laughlin state~\cite{Laughlin1983} with incoherent errors.
%
In this case, we find two phases summarized in Table~I.
The low-energy excitations in the Laughlin state is described by the abelian CS theory with K-matrix $K_{1/3} = (3)$.
%
It contains three superselection sectors $\{1, \eta, \eta^2\}$.
%
$\eta$ and $\eta^2$ are the quasiparticle and quasihole, and they are labeled by the charge vector $\bm{l}_\eta = (1)$ and $\bm{l}_{\eta^2} = (2)$ and carry charge $e/3$ and $-e/3$, respectively.
There is only one type of incoherent channel; its Kraus operators create a pair of quasiparticle and quasihole.
%
When the error is below the threshold, the corrupted state is a quantum memory characterized by the Lagrangian subgroup generated by $\eta_L \eta_R^2$ and $\bar{\eta}_L\bar{\eta}_R^2$ (Phase I in Table~I).
%
When error proliferates, we obtain a trivial phase with a subgroup generated by $\eta_L \bar{\eta}_L$ and $\eta_R \bar{\eta}_R$ (Phase II in Table~I).
\subsection{Generalization to the $n$-th moment}\label{sec:nth}
\begin{figure}
\centering
\begin{tikzpicture}
\pgftext{\includegraphics[width=0.7\textwidth]{suppfig_nth.pdf}} at (0pt,0pt);
%
%
\node at (-5.5, 0.96) {(a)};
\draw [->,>=stealth] (-3.6, 0.7) node[above]{1D defect} -- (-3.7,0.4);
\node at (-3.6, -0.9) {2 $\times$ (topo. order)};
%
\node at (-1.0, 0.96) {(b)};
\node at (-1.5, 0.05) {\scriptsize Folding};
\node at (0.3, 0.4) {\scriptsize $\phi_1$};
\node at (0.3, 0) {\scriptsize $\bar{\phi}_1$};
\node at (0.3, -0.4) {\scriptsize $\phi_2$};
\node at (0.3, -0.8) {\scriptsize $\bar{\phi}_2$};
%
\node at (2.3, 1.4) {(c)};
\node at (3.5, 0.85) {\scriptsize $\phi_1$};
\node at (3.5, 0.5) {\scriptsize $\bar{\phi}_1$};
\node at (3.5, 0.05) {\scriptsize $\phi_2$};
\node at (3.5, -0.3) {\scriptsize $\bar{\phi}_2$};
\node at (4.9, -0.7) {\scriptsize $\vdots$};
\node at (3.5, -0.93) {\scriptsize $\phi_n$};
\node at (3.5, -1.28) {\scriptsize $\bar{\phi}_n$};
\end{tikzpicture}
\caption{(a) The path integral represents the norm of the ground state of double 2D topological order with 1D defect (red region). (b) Folding the double topological order around the 1D defect results in quadruple topological order with a 1D boundary. The red regions represent the coupling $\calL_\calN$ induced by the quantum channel. The couplings between $\phi_1$ and $\bar{\phi}_2$ (also $\bar{\phi}_1$ and $\phi_2$) are denoted by $\calL_1$ inherited from the definition of path integral. (c) The edge theory associated with the $n$-th moment $\tr\rho^n$ of the density matrix. The generalized swap interaction $\calL_1^{(n)}$ couples the compact bosons in the neighboring copies, i.e. $\bar{\phi}_{s}$ and $\phi_{s+1}$ for $s = 1, 2,\cdots, n$.}
\label{suppfig:nth_moment}
\end{figure}
So far, the analysis focuses on the distinct phases in the EFD, which pertains to analyzing the second moment of the density matrix, i.e. the norm of the EFD $\bbrakket{\rho}{\rho} = \tr \rho^2$.
%
In this subsection, we generalize the problem to the $n$-th moment $\tr\rho^n$ and map the decoherence-induced phases to $(1+1)$D boundary phases.
Before discussing the $n$-th moment, we develop an equivalent formulation of the second moment $\tr \rho^2$.
%
The phases of the $(1+1)$D defect can be converted to the $(1+1)$D edge phases by folding around the defect as shown in Fig.~\ref{suppfig:nth_moment}(b)~\cite{BeigiShor2011,Kapustin2011}.
%
In the folded picture, we have four copies of the topological order.
%
On the edge, we relabel the bosonic fields as
\begin{align}
\phi := [\bar{\phi}_R, \phi_R, \phi_L, \bar{\phi}_L] = [\phi_1, \bar{\phi}_1, \phi_2, \bar{\phi}_2],
\end{align}
where $\phi_{1(2)}$ and $\bar{\phi}_{1(2)}$ denote the fields associated with the copy-$1(2)$ of the density matrix.
The path integral describing the ground state of quadruple topological order with an edge corresponds to an overlap in the quadruple Hilbert space $\calH^{\otimes 4}$ before the rotation
\begin{align}
\tr\rho^2 = \bbra{\mathbb{C}^{(2)}}\left(\kket{\rho}\otimes \kket{\rho}\right) := \int \mathcal{D}(\varphi_1, \varphi_2) \left(\bra{\varphi_1}\otimes \bra{\varphi_2}\right)\otimes \kket{\rho} \, \left(\bra{\varphi_2}\otimes \bra{\varphi_1}\right)\otimes \kket{\rho},
\end{align}
where $\bbra{\mathbb{C}^{(2)}}$ is a ``swap" reference state that couples the fields in the ket and bra Hilbert space of the different copies of the EFD.
The overlap is readily generalized to describe the $n$-th moment $\tr\rho^n$.
%
Here, in $2n$ copies of the Hilbert space $\calH^{\otimes 2n}$, we consider
\begin{align}
\tr\rho^n = \bbra{\mathbb{C}^{(n)}}\left(\kket{\rho}^{\otimes n}\right) := \int \mathcal{D}(\varphi_1, \varphi_2, \cdots, \varphi_n) \prod_{s = 1}^n \left(\bra{\varphi_s}\otimes \bra{\varphi_{s+1}}\right)\otimes \kket{\rho},
\end{align}
where $\varphi_{n+1} := \varphi_1$, and $\bbra{\mathbb{C}^{(n)}}$ is a generalized swap reference state.
We can similarly develop a path integral formulation of the overlap.
%
In the rotated picture, it corresponds to the ground state of $2n$ copies of the topological order with an $(1+1)$D edge.
%
The effective edge theory also contains three parts
\begin{equation}
\calL^{(n)} = \calL_0^{(n)} + \calL_1^{(n)} + \calL_\calN^{(n)}.
\end{equation}
The first part $\calL_0^{(n)}$ describes the low-energy excitations in $2n$ copies of the topological order
\begin{align}
\calL_0^{(n)}[\phi] = \frac{1}{4\pi} \sum_{I,J} \mathbb{K}_{IJ}^{(n)} \ri\partial_\tau \phi^I \partial_y \phi^J - \mathbb{V}_{IJ}^{(n)} \partial_y \phi^I \partial_y \phi^J
\end{align}
where the compact bosonic field $\phi := [\phi_1, \bar{\phi}_1, \phi_2, \bar{\phi}_2, \cdots, \phi_n, \bar{\phi}_n]$ is a $2nM$-component vector, $\mathbb{K}^{(n)} = [K \oplus (-K)]^{\oplus n}$ , and $\mathbb{V}^{(n)} = V^{\oplus 2n}$.
%
The second part $\calL_1^{(n)}$ is the interaction introduced by the reference state, which takes a general form
\begin{align}
\calL_1^{(n)} = \sum_\Lambda B_\Lambda \sum_{s = 1}^n \cos\left( \sum_I (K\Lambda)_I (\phi_s^I + \bar{\phi}_{s+1}^I) + \theta \right).
\end{align}
where $\bar{\phi}_{n+1} := \bar{\phi}_1$.
%
The last part $\calL_\calN^{(n)}$ describes the error channel. For incoherent errors, we have
\begin{align}
\calL_\calN^{(n)} = \sum_\Lambda C_\Lambda \sum_{s = 1}^n \cos\left( \sum_I (K\Lambda)_I (\phi_s^I - \bar{\phi}_s^I)\right).
\end{align}
The edge theory exhibits a global symmetry $\mathcal{G}^{(n)} = \mathbb{Z}_n \times \mathbb{Z}_2^\mathbb{H}$.
%
We note that the $\mathbb{Z}_n$ symmetry is a unitary symmetry generated by the cyclic permutation over $n$ copies of the density matrix
\begin{align}
\phi_s \leftrightarrow \phi_{s+1}, \quad \bar{\phi}_s \leftrightarrow \bar{\phi}_{s+1}.
\end{align}
In the case of $n = 2$, $\calG^{(n)}$ reduces to $\calG^{(2)}$ in Sec.~\ref{sec:mapping} as the permutation over two copies of the density matrix is a combination of the Hermitian conjugation of the EFD and that of the density matrix.
We remark that the incoherent errors can induce condensation of $\alpha_s \bar{\alpha}_s$ for $s = 1, 2, \cdots n$ on the edge.
%
Since the different anyon bound states $\alpha_s \bar{\alpha}_s$ and $\beta_{s'}\bar{\beta}_{s'}$ are mutual bosons, different incoherent channels do not compete with each other.
%
Thus, the number of possible condensates is independent of the replica index $n$, and so is the classification for abelian topological with incoherent errors.
%
Hence, we expect each phase of $\tr\rho^2$ in Table~I has a correspondence in the replica limit $n \to 1$.
\bibliography{refs_short.bib}
\end{document}
%apsrev4-2.bst 2019-01-14 (MD) hand-edited version of apsrev4-1.bst
%Control: key (0)
%Control: author (8) initials jnrlst
%Control: editor formatted (1) identically to author
%Control: production of article title (0) allowed
%Control: page (0) single
%Control: year (1) truncated
%Control: production of eprint (0) enabled
\begin{thebibliography}{63}%
\makeatletter
\providecommand \@ifxundefined [1]{%
\@ifx{#1\undefined}
}%
\providecommand \@ifnum [1]{%
\ifnum #1\expandafter \@firstoftwo
\else \expandafter \@secondoftwo
\fi
}%
\providecommand \@ifx [1]{%
\ifx #1\expandafter \@firstoftwo
\else \expandafter \@secondoftwo
\fi
}%
\providecommand \natexlab [1]{#1}%
\providecommand \enquote [1]{``#1''}%
\providecommand \bibnamefont [1]{#1}%
\providecommand \bibfnamefont [1]{#1}%
\providecommand \citenamefont [1]{#1}%
\providecommand \href@noop [0]{\@secondoftwo}%
\providecommand \href [0]{\begingroup \@sanitize@url \@href}%
\providecommand \@href[1]{\@@startlink{#1}\@@href}%
\providecommand \@@href[1]{\endgroup#1\@@endlink}%
\providecommand \@sanitize@url [0]{\catcode `\\12\catcode `\$12\catcode
`\&12\catcode `\#12\catcode `\^12\catcode `\_12\catcode `\%12\relax}%
\providecommand \@@startlink[1]{}%
\providecommand \@@endlink[0]{}%
\providecommand \url [0]{\begingroup\@sanitize@url \@url }%
\providecommand \@url [1]{\endgroup\@href {#1}{\urlprefix }}%
\providecommand \urlprefix [0]{URL }%
\providecommand \Eprint [0]{\href }%
\providecommand \doibase [0]{https://doi.org/}%
\providecommand \selectlanguage [0]{\@gobble}%
\providecommand \bibinfo [0]{\@secondoftwo}%
\providecommand \bibfield [0]{\@secondoftwo}%
\providecommand \translation [1]{[#1]}%
\providecommand \BibitemOpen [0]{}%
\providecommand \bibitemStop [0]{}%
\providecommand \bibitemNoStop [0]{.\EOS\space}%
\providecommand \EOS [0]{\spacefactor3000\relax}%
\providecommand \BibitemShut [1]{\csname bibitem#1\endcsname}%
\let\auto@bib@innerbib\@empty
%</preamble>
\bibitem [{\citenamefont {Semeghini}\ \emph {et~al.}(2021)\citenamefont
{Semeghini} \emph {et~al.}}]{Semeghini:2021wls}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {G.}~\bibnamefont
{Semeghini}} \emph {et~al.},\ }\bibfield {title} {\bibinfo {title} {{Probing
topological spin liquids on a programmable quantum simulator}},\ }\href
{https://doi.org/10.1126/science.abi8794} {\bibfield {journal} {\bibinfo
{journal} {Science}\ }\textbf {\bibinfo {volume} {374}},\ \bibinfo {pages}
{abi8794} (\bibinfo {year} {2021})},\ \Eprint
{https://arxiv.org/abs/2104.04119} {arXiv:2104.04119 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Bluvstein}\ \emph {et~al.}(2022)\citenamefont
{Bluvstein} \emph {et~al.}}]{Bluvstein:2021jsq}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {D.}~\bibnamefont
{Bluvstein}} \emph {et~al.},\ }\bibfield {title} {\bibinfo {title} {{A
quantum processor based on coherent transport of entangled atom arrays}},\
}\href {https://doi.org/10.1038/s41586-022-04592-6} {\bibfield {journal}
{\bibinfo {journal} {Nature}\ }\textbf {\bibinfo {volume} {604}},\ \bibinfo
{pages} {451} (\bibinfo {year} {2022})},\ \Eprint
{https://arxiv.org/abs/2112.03923} {arXiv:2112.03923 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Satzinger}\ \emph {et~al.}(2021)\citenamefont
{Satzinger} \emph {et~al.}}]{Satzinger:2021eqy}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {K.~J.}\ \bibnamefont
{Satzinger}} \emph {et~al.},\ }\bibfield {title} {\bibinfo {title}
{{Realizing topologically ordered states on a quantum processor}},\ }\href
{https://doi.org/10.1126/science.abi8378} {\bibfield {journal} {\bibinfo
{journal} {Science}\ }\textbf {\bibinfo {volume} {374}},\ \bibinfo {pages}
{abi8378} (\bibinfo {year} {2021})},\ \Eprint
{https://arxiv.org/abs/2104.01180} {arXiv:2104.01180 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Acharya}\ \emph {et~al.}(2022)\citenamefont {Acharya}
\emph {et~al.}}]{GoogleQuantumAI:2022fyn}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {R.}~\bibnamefont
{Acharya}} \emph {et~al.} (\bibinfo {collaboration} {Google Quantum AI}),\
}\bibfield {title} {\bibinfo {title} {{Suppressing quantum errors by scaling
a surface code logical qubit}},\ }\href@noop {} {\ (\bibinfo {year}
{2022})},\ \Eprint {https://arxiv.org/abs/2207.06431} {arXiv:2207.06431
[quant-ph]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {Tsui}\ \emph {et~al.}(1982)\citenamefont {Tsui},
\citenamefont {Stormer},\ and\ \citenamefont {Gossard}}]{Tsui:1982yy}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {D.~C.}\ \bibnamefont
{Tsui}}, \bibinfo {author} {\bibfnamefont {H.~L.}\ \bibnamefont {Stormer}},\
and\ \bibinfo {author} {\bibfnamefont {A.~C.}\ \bibnamefont {Gossard}},\
}\bibfield {title} {\bibinfo {title} {{Two-dimensional magnetotransport in
the extreme quantum limit}},\ }\href
{https://doi.org/10.1103/PhysRevLett.48.1559} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. Lett.}\ }\textbf {\bibinfo {volume} {48}},\ \bibinfo
{pages} {1559} (\bibinfo {year} {1982})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Bartolomei}}\ \emph {et~al.}(2020)\citenamefont
{{Bartolomei}}, \citenamefont {{Kumar}}, \citenamefont {{Bisognin}},
\citenamefont {{Marguerite}}, \citenamefont {{Berroir}}, \citenamefont
{{Bocquillon}}, \citenamefont {{Pla{\c{c}}ais}}, \citenamefont {{Cavanna}},
\citenamefont {{Dong}}, \citenamefont {{Gennser}}, \citenamefont {{Jin}},\
and\ \citenamefont {{F{\`e}ve}}}]{AnyonCollisions}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {H.}~\bibnamefont
{{Bartolomei}}}, \bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{{Kumar}}}, \bibinfo {author} {\bibfnamefont {R.}~\bibnamefont {{Bisognin}}},
\bibinfo {author} {\bibfnamefont {A.}~\bibnamefont {{Marguerite}}}, \bibinfo
{author} {\bibfnamefont {J.~M.}\ \bibnamefont {{Berroir}}}, \bibinfo {author}
{\bibfnamefont {E.}~\bibnamefont {{Bocquillon}}}, \bibinfo {author}
{\bibfnamefont {B.}~\bibnamefont {{Pla{\c{c}}ais}}}, \bibinfo {author}
{\bibfnamefont {A.}~\bibnamefont {{Cavanna}}}, \bibinfo {author}
{\bibfnamefont {Q.}~\bibnamefont {{Dong}}}, \bibinfo {author} {\bibfnamefont
{U.}~\bibnamefont {{Gennser}}}, \bibinfo {author} {\bibfnamefont
{Y.}~\bibnamefont {{Jin}}},\ and\ \bibinfo {author} {\bibfnamefont
{G.}~\bibnamefont {{F{\`e}ve}}},\ }\bibfield {title} {\bibinfo {title}
{{Fractional statistics in anyon collisions}},\ }\href
{https://doi.org/10.1126/science.aaz5601} {\bibfield {journal} {\bibinfo
{journal} {Science}\ }\textbf {\bibinfo {volume} {368}},\ \bibinfo {pages}
{173} (\bibinfo {year} {2020})},\ \Eprint {https://arxiv.org/abs/2006.13157}
{arXiv:2006.13157 [cond-mat.mes-hall]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Nakamura}}\ \emph {et~al.}(2020)\citenamefont
{{Nakamura}}, \citenamefont {{Liang}}, \citenamefont {{Gardner}},\ and\
\citenamefont {{Manfra}}}]{AnyonBraid}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {J.}~\bibnamefont
{{Nakamura}}}, \bibinfo {author} {\bibfnamefont {S.}~\bibnamefont {{Liang}}},
\bibinfo {author} {\bibfnamefont {G.~C.}\ \bibnamefont {{Gardner}}},\ and\
\bibinfo {author} {\bibfnamefont {M.~J.}\ \bibnamefont {{Manfra}}},\
}\bibfield {title} {\bibinfo {title} {{Direct observation of anyonic
braiding statistics}},\ }\href {https://doi.org/10.1038/s41567-020-1019-1}
{\bibfield {journal} {\bibinfo {journal} {Nature Physics}\ }\textbf
{\bibinfo {volume} {16}},\ \bibinfo {pages} {931} (\bibinfo {year} {2020})},\
\Eprint {https://arxiv.org/abs/2006.14115} {arXiv:2006.14115
[cond-mat.mes-hall]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Wen}}(2017)}]{xiaogangReview}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {X.-G.}\ \bibnamefont
{{Wen}}},\ }\bibfield {title} {\bibinfo {title} {{Colloquium: Zoo of
quantum-topological phases of matter}},\ }\href
{https://doi.org/10.1103/RevModPhys.89.041004} {\bibfield {journal}
{\bibinfo {journal} {Reviews of Modern Physics}\ }\textbf {\bibinfo {volume}
{89}},\ \bibinfo {eid} {041004} (\bibinfo {year} {2017})},\ \Eprint
{https://arxiv.org/abs/1610.03911} {arXiv:1610.03911 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Nussinov}}\ and\ \citenamefont
{{Ortiz}}(2008)}]{Nussinov2008PRB}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {Z.}~\bibnamefont
{{Nussinov}}}\ and\ \bibinfo {author} {\bibfnamefont {G.}~\bibnamefont
{{Ortiz}}},\ }\bibfield {title} {\bibinfo {title} {{Autocorrelations and
thermal fragility of anyonic loops in topologically quantum ordered
systems}},\ }\href {https://doi.org/10.1103/PhysRevB.77.064302} {\bibfield
{journal} {\bibinfo {journal} {\prb}\ }\textbf {\bibinfo {volume} {77}},\
\bibinfo {eid} {064302} (\bibinfo {year} {2008})},\ \Eprint
{https://arxiv.org/abs/0709.2717} {arXiv:0709.2717 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Nussinov}}\ and\ \citenamefont
{{Ortiz}}(2009)}]{Nussinov2009PNAS}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {Z.}~\bibnamefont
{{Nussinov}}}\ and\ \bibinfo {author} {\bibfnamefont {G.}~\bibnamefont
{{Ortiz}}},\ }\bibfield {title} {\bibinfo {title} {{Sufficient symmetry
conditions for Topological Quantum Order}},\ }\href
{https://doi.org/10.1073/pnas.0803726105} {\bibfield {journal} {\bibinfo
{journal} {Proceedings of the National Academy of Science}\ }\textbf
{\bibinfo {volume} {106}},\ \bibinfo {pages} {16944} (\bibinfo {year}
{2009})},\ \Eprint {https://arxiv.org/abs/cond-mat/0605316}
{arXiv:cond-mat/0605316 [cond-mat.str-el]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Bravyi}}\ and\ \citenamefont
{{Terhal}}(2009)}]{Bravyi2009NoGo}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {S.}~\bibnamefont
{{Bravyi}}}\ and\ \bibinfo {author} {\bibfnamefont {B.}~\bibnamefont
{{Terhal}}},\ }\bibfield {title} {\bibinfo {title} {{A no-go theorem for a
two-dimensional self-correcting quantum memory based on stabilizer codes}},\
}\href {https://doi.org/10.1088/1367-2630/11/4/043029} {\bibfield {journal}
{\bibinfo {journal} {New Journal of Physics}\ }\textbf {\bibinfo {volume}
{11}},\ \bibinfo {eid} {043029} (\bibinfo {year} {2009})},\ \Eprint
{https://arxiv.org/abs/0810.1983} {arXiv:0810.1983 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {{Landon-Cardinal}}\ and\ \citenamefont
{{Poulin}}(2013)}]{Poulin2013PRL}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {O.}~\bibnamefont
{{Landon-Cardinal}}}\ and\ \bibinfo {author} {\bibfnamefont {D.}~\bibnamefont
{{Poulin}}},\ }\bibfield {title} {\bibinfo {title} {{Local Topological Order
Inhibits Thermal Stability in 2D}},\ }\href
{https://doi.org/10.1103/PhysRevLett.110.090502} {\bibfield {journal}
{\bibinfo {journal} {\prl}\ }\textbf {\bibinfo {volume} {110}},\ \bibinfo
{eid} {090502} (\bibinfo {year} {2013})},\ \Eprint
{https://arxiv.org/abs/1209.5750} {arXiv:1209.5750 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {{Brown}}\ \emph {et~al.}(2016)\citenamefont
{{Brown}}, \citenamefont {{Loss}}, \citenamefont {{Pachos}}, \citenamefont
{{Self}},\ and\ \citenamefont {{Wootton}}}]{Brown2016RMP}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {B.~J.}\ \bibnamefont
{{Brown}}}, \bibinfo {author} {\bibfnamefont {D.}~\bibnamefont {{Loss}}},
\bibinfo {author} {\bibfnamefont {J.~K.}\ \bibnamefont {{Pachos}}}, \bibinfo
{author} {\bibfnamefont {C.~N.}\ \bibnamefont {{Self}}},\ and\ \bibinfo
{author} {\bibfnamefont {J.~R.}\ \bibnamefont {{Wootton}}},\ }\bibfield
{title} {\bibinfo {title} {{Quantum memories at finite temperature}},\ }\href
{https://doi.org/10.1103/RevModPhys.88.045005} {\bibfield {journal}
{\bibinfo {journal} {Reviews of Modern Physics}\ }\textbf {\bibinfo {volume}
{88}},\ \bibinfo {eid} {045005} (\bibinfo {year} {2016})},\ \Eprint
{https://arxiv.org/abs/1411.6643} {arXiv:1411.6643 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Fradkin}(2013)}]{Fradkin:2013sab}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {E.~H.}\ \bibnamefont
{Fradkin}},\ }\href@noop {} {\emph {\bibinfo {title} {{Field Theories of
Condensed Matter Physics}}}},\ Vol.~\bibinfo {volume} {82}\ (\bibinfo
{publisher} {Cambridge Univ. Press},\ \bibinfo {address} {Cambridge, UK},\
\bibinfo {year} {2013})\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Kitaev}(2003)}]{Kitaev:1997wr}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {A.~Y.}\ \bibnamefont
{Kitaev}},\ }\bibfield {title} {\bibinfo {title} {{Fault tolerant quantum
computation by anyons}},\ }\href
{https://doi.org/10.1016/S0003-4916(02)00018-0} {\bibfield {journal}
{\bibinfo {journal} {Annals Phys.}\ }\textbf {\bibinfo {volume} {303}},\
\bibinfo {pages} {2} (\bibinfo {year} {2003})},\ \Eprint
{https://arxiv.org/abs/quant-ph/9707021} {arXiv:quant-ph/9707021}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Bravyi}\ and\ \citenamefont
{Kitaev}(1998)}]{Bravyi:1998sy}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {S.~B.}\ \bibnamefont
{Bravyi}}\ and\ \bibinfo {author} {\bibfnamefont {A.~Y.}\ \bibnamefont
{Kitaev}},\ }\bibfield {title} {\bibinfo {title} {{Quantum codes on a
lattice with boundary}},\ }\href@noop {} {\ (\bibinfo {year} {1998})},\
\Eprint {https://arxiv.org/abs/quant-ph/9811052} {arXiv:quant-ph/9811052}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Dennis}\ \emph {et~al.}(2002)\citenamefont {Dennis},
\citenamefont {Kitaev}, \citenamefont {Landahl},\ and\ \citenamefont
{Preskill}}]{Dennis:2001nw}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {E.}~\bibnamefont
{Dennis}}, \bibinfo {author} {\bibfnamefont {A.}~\bibnamefont {Kitaev}},
\bibinfo {author} {\bibfnamefont {A.}~\bibnamefont {Landahl}},\ and\ \bibinfo
{author} {\bibfnamefont {J.}~\bibnamefont {Preskill}},\ }\bibfield {title}
{\bibinfo {title} {{Topological quantum memory}},\ }\href
{https://doi.org/10.1063/1.1499754} {\bibfield {journal} {\bibinfo
{journal} {J. Math. Phys.}\ }\textbf {\bibinfo {volume} {43}},\ \bibinfo
{pages} {4452} (\bibinfo {year} {2002})},\ \Eprint
{https://arxiv.org/abs/quant-ph/0110143} {arXiv:quant-ph/0110143}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Garratt}\ \emph {et~al.}(2022)\citenamefont
{Garratt}, \citenamefont {Weinstein},\ and\ \citenamefont
{Altman}}]{Garratt:2022ycp}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {S.~J.}\ \bibnamefont
{Garratt}}, \bibinfo {author} {\bibfnamefont {Z.}~\bibnamefont {Weinstein}},\
and\ \bibinfo {author} {\bibfnamefont {E.}~\bibnamefont {Altman}},\
}\bibfield {title} {\bibinfo {title} {{Measurements conspire nonlocally to
restructure critical quantum states}},\ }\href@noop {} {\ (\bibinfo {year}
{2022})},\ \Eprint {https://arxiv.org/abs/2207.09476} {arXiv:2207.09476
[cond-mat.stat-mech]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {Levin}(2013)}]{Levin:2013gaa}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Levin}},\ }\bibfield {title} {\bibinfo {title} {{Protected edge modes
without symmetry}},\ }\href {https://doi.org/10.1103/PhysRevX.3.021009}
{\bibfield {journal} {\bibinfo {journal} {Phys. Rev. X}\ }\textbf {\bibinfo
{volume} {3}},\ \bibinfo {pages} {021009} (\bibinfo {year} {2013})},\ \Eprint
{https://arxiv.org/abs/1301.7355} {arXiv:1301.7355 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Barkeshli}\ \emph {et~al.}(2013)\citenamefont
{Barkeshli}, \citenamefont {Jian},\ and\ \citenamefont
{Qi}}]{Barkeshli:2013}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Barkeshli}}, \bibinfo {author} {\bibfnamefont {C.-M.}\ \bibnamefont
{Jian}},\ and\ \bibinfo {author} {\bibfnamefont {X.-L.}\ \bibnamefont {Qi}},\
}\bibfield {title} {\bibinfo {title} {{Theory of defects in Abelian
topological states}},\ }\href {https://doi.org/10.1103/PhysRevB.88.235103}
{\bibfield {journal} {\bibinfo {journal} {Phys. Rev. B}\ }\textbf {\bibinfo
{volume} {88}},\ \bibinfo {pages} {235103} (\bibinfo {year} {2013})},\
\Eprint {https://arxiv.org/abs/1305.7203} {arXiv:1305.7203 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Wang}\ and\ \citenamefont
{Preskill}(2003)}]{Wang:2002ph}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {C.}~\bibnamefont
{Wang}}\ and\ \bibinfo {author} {\bibfnamefont {J.}~\bibnamefont
{Preskill}},\ }\bibfield {title} {\bibinfo {title} {{Confinement Higgs
transition in a disordered gauge theory and the accuracy threshold for
quantum memory}},\ }\href {https://doi.org/10.1016/S0003-4916(02)00019-2}
{\bibfield {journal} {\bibinfo {journal} {Annals Phys.}\ }\textbf {\bibinfo
{volume} {303}},\ \bibinfo {pages} {31} (\bibinfo {year} {2003})},\ \Eprint
{https://arxiv.org/abs/quant-ph/0207088} {arXiv:quant-ph/0207088}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Katzgraber}\ \emph {et~al.}(2009)\citenamefont
{Katzgraber}, \citenamefont {Bombin},\ and\ \citenamefont
{Martin-Delgado}}]{Katzgraber:2009zz}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {H.~G.}\ \bibnamefont
{Katzgraber}}, \bibinfo {author} {\bibfnamefont {H.}~\bibnamefont {Bombin}},\
and\ \bibinfo {author} {\bibfnamefont {M.~A.}\ \bibnamefont
{Martin-Delgado}},\ }\bibfield {title} {\bibinfo {title} {{Error Threshold
for Color Codes and Random Three-Body Ising Models}},\ }\href
{https://doi.org/10.1103/PhysRevLett.103.090501} {\bibfield {journal}
{\bibinfo {journal} {Phys. Rev. Lett.}\ }\textbf {\bibinfo {volume} {103}},\
\bibinfo {pages} {090501} (\bibinfo {year} {2009})},\ \Eprint
{https://arxiv.org/abs/0902.4845} {arXiv:0902.4845 [cond-mat.dis-nn]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Bombin}\ \emph {et~al.}(2012)\citenamefont {Bombin},
\citenamefont {Andrist}, \citenamefont {Ohzeki}, \citenamefont {Katzgraber},\
and\ \citenamefont {Martin-Delgado}}]{Bombin:2012jk}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {H.}~\bibnamefont
{Bombin}}, \bibinfo {author} {\bibfnamefont {R.~S.}\ \bibnamefont {Andrist}},
\bibinfo {author} {\bibfnamefont {M.}~\bibnamefont {Ohzeki}}, \bibinfo
{author} {\bibfnamefont {H.~G.}\ \bibnamefont {Katzgraber}},\ and\ \bibinfo
{author} {\bibfnamefont {M.~A.}\ \bibnamefont {Martin-Delgado}},\ }\bibfield
{title} {\bibinfo {title} {{Strong resilience of topological codes to
depolarization}},\ }\href {https://doi.org/10.1103/PhysRevX.2.021004}
{\bibfield {journal} {\bibinfo {journal} {Phys. Rev. X}\ }\textbf {\bibinfo
{volume} {2}},\ \bibinfo {pages} {021004} (\bibinfo {year} {2012})},\ \Eprint
{https://arxiv.org/abs/1202.1852} {arXiv:1202.1852 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Kubica}\ \emph {et~al.}(2018)\citenamefont {Kubica},
\citenamefont {Beverland}, \citenamefont {Brand\~ao}, \citenamefont
{Preskill},\ and\ \citenamefont {Svore}}]{Kubica:2018rab}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {A.}~\bibnamefont
{Kubica}}, \bibinfo {author} {\bibfnamefont {M.~E.}\ \bibnamefont
{Beverland}}, \bibinfo {author} {\bibfnamefont {F.}~\bibnamefont
{Brand\~ao}}, \bibinfo {author} {\bibfnamefont {J.}~\bibnamefont
{Preskill}},\ and\ \bibinfo {author} {\bibfnamefont {K.~M.}\ \bibnamefont
{Svore}},\ }\bibfield {title} {\bibinfo {title} {{Three-Dimensional Color
Code Thresholds via Statistical-Mechanical Mapping}},\ }\href
{https://doi.org/10.1103/PhysRevLett.120.180501} {\bibfield {journal}
{\bibinfo {journal} {Phys. Rev. Lett.}\ }\textbf {\bibinfo {volume} {120}},\
\bibinfo {pages} {180501} (\bibinfo {year} {2018})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Chubb}\ and\ \citenamefont
{Flammia}(2021)}]{Chubb:2021htd}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {C.~T.}\ \bibnamefont
{Chubb}}\ and\ \bibinfo {author} {\bibfnamefont {S.~T.}\ \bibnamefont
{Flammia}},\ }\bibfield {title} {\bibinfo {title} {{Statistical mechanical
models for quantum codes with correlated noise}},\ }\href
{https://doi.org/10.4171/aihpd/105} {\bibfield {journal} {\bibinfo
{journal} {Ann. Inst. H. Poincare D Comb. Phys. Interact.}\ }\textbf
{\bibinfo {volume} {8}},\ \bibinfo {pages} {269} (\bibinfo {year}
{2021})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Castelnovo}}\ and\ \citenamefont
{{Chamon}}(2008)}]{CastelnovoPRB2008}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {C.}~\bibnamefont
{{Castelnovo}}}\ and\ \bibinfo {author} {\bibfnamefont {C.}~\bibnamefont
{{Chamon}}},\ }\bibfield {title} {\bibinfo {title} {{Quantum topological
phase transition at the microscopic level}},\ }\href
{https://doi.org/10.1103/PhysRevB.77.054433} {\bibfield {journal} {\bibinfo
{journal} {\prb}\ }\textbf {\bibinfo {volume} {77}},\ \bibinfo {eid} {054433}
(\bibinfo {year} {2008})},\ \Eprint {https://arxiv.org/abs/0707.2084}
{arXiv:0707.2084 [cond-mat.str-el]} \BibitemShut {NoStop}%
\bibitem [{Note1()}]{Note1}%
\BibitemOpen
\bibinfo {note} {Strictly speaking, such a decomposition is well-defined for
fixed-point states where the sizes of anyons are zero. In more general
situations, e.g., Laughlin states, one can regard this decomposition as a
\protect \emph {definition} of phenomenological error models that we choose
to work with.}\BibitemShut {Stop}%
\bibitem [{\citenamefont {Haag}(1992)}]{Haag:1992hx}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {R.}~\bibnamefont
{Haag}},\ }\href@noop {} {\emph {\bibinfo {title} {{Local quantum physics:
Fields, particles, algebras}}}}\ (\bibinfo {year} {1992})\BibitemShut
{NoStop}%
\bibitem [{SOM()}]{SOM}%
\BibitemOpen
\href@noop {} {}\bibinfo {note} {See supplementary online material for
details.}\BibitemShut {Stop}%
\bibitem [{Note2()}]{Note2}%
\BibitemOpen
\bibinfo {note} {The superscript of $p_c^{(2)}$ indicates the transition in
EFD is detected by the quadratic function of the density matrix. For the
general $n$-th moment of the density matrix, the critical point $p_c^{(n)}$
depends on $n$. It approaches the error-correction threshold $p_c = 0.109$ in
the replica limit $n \to 1$~\cite {Diagnosticstoappear}.}\BibitemShut {Stop}%
\bibitem [{\citenamefont {Fredenhagen}\ and\ \citenamefont
{Marcu}(1983)}]{Fredenhagen:1983sn}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {K.}~\bibnamefont
{Fredenhagen}}\ and\ \bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Marcu}},\ }\bibfield {title} {\bibinfo {title} {{Charged States in Z2 Gauge
Theories}},\ }\href {https://doi.org/10.1007/BF01206315} {\bibfield
{journal} {\bibinfo {journal} {Commun. Math. Phys.}\ }\textbf {\bibinfo
{volume} {92}},\ \bibinfo {pages} {81} (\bibinfo {year} {1983})}\BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Fredenhagen}\ and\ \citenamefont
{Marcu}(1986)}]{Fredenhagen:1985ft}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {K.}~\bibnamefont
{Fredenhagen}}\ and\ \bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Marcu}},\ }\bibfield {title} {\bibinfo {title} {{A Confinement Criterion
for {QCD} With Dynamical Quarks}},\ }\href
{https://doi.org/10.1103/PhysRevLett.56.223} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. Lett.}\ }\textbf {\bibinfo {volume} {56}},\ \bibinfo
{pages} {223} (\bibinfo {year} {1986})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Fredenhagen}\ and\ \citenamefont
{Marcu}(1988)}]{Fredenhagen:1987gi}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {K.}~\bibnamefont
{Fredenhagen}}\ and\ \bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Marcu}},\ }\bibfield {title} {\bibinfo {title} {{DUAL INTERPRETATION OF
ORDER PARAMETERS FOR LATTICE GAUGE THEORIES WITH MATTER FIELDS}},\ }\href
{https://doi.org/10.1016/0920-5632(88)90124-7} {\bibfield {journal}
{\bibinfo {journal} {Nucl. Phys. B Proc. Suppl.}\ }\textbf {\bibinfo
{volume} {4}},\ \bibinfo {pages} {352} (\bibinfo {year} {1988})}\BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Gregor}\ \emph {et~al.}(2011)\citenamefont {Gregor},
\citenamefont {Huse}, \citenamefont {Moessner},\ and\ \citenamefont
{Sondhi}}]{Gregor:2010ym}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {K.}~\bibnamefont
{Gregor}}, \bibinfo {author} {\bibfnamefont {D.~A.}\ \bibnamefont {Huse}},
\bibinfo {author} {\bibfnamefont {R.}~\bibnamefont {Moessner}},\ and\
\bibinfo {author} {\bibfnamefont {S.~L.}\ \bibnamefont {Sondhi}},\ }\bibfield
{title} {\bibinfo {title} {{Diagnosing Deconfinement and Topological
Order}},\ }\href {https://doi.org/10.1088/1367-2630/13/2/025009} {\bibfield
{journal} {\bibinfo {journal} {New J. Phys.}\ }\textbf {\bibinfo {volume}
{13}},\ \bibinfo {pages} {025009} (\bibinfo {year} {2011})},\ \Eprint
{https://arxiv.org/abs/1011.4187} {arXiv:1011.4187 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Verresen}\ \emph {et~al.}(2021)\citenamefont
{Verresen}, \citenamefont {Lukin},\ and\ \citenamefont
{Vishwanath}}]{Verresen:2020dmk}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {R.}~\bibnamefont
{Verresen}}, \bibinfo {author} {\bibfnamefont {M.~D.}\ \bibnamefont
{Lukin}},\ and\ \bibinfo {author} {\bibfnamefont {A.}~\bibnamefont
{Vishwanath}},\ }\bibfield {title} {\bibinfo {title} {{Prediction of Toric
Code Topological Order from Rydberg Blockade}},\ }\href
{https://doi.org/10.1103/PhysRevX.11.031005} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. X}\ }\textbf {\bibinfo {volume} {11}},\ \bibinfo
{pages} {031005} (\bibinfo {year} {2021})},\ \Eprint
{https://arxiv.org/abs/2011.12310} {arXiv:2011.12310 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Kitaev}\ and\ \citenamefont
{Preskill}(2006)}]{Kitaev:2005dm}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {A.}~\bibnamefont
{Kitaev}}\ and\ \bibinfo {author} {\bibfnamefont {J.}~\bibnamefont
{Preskill}},\ }\bibfield {title} {\bibinfo {title} {{Topological
entanglement entropy}},\ }\href
{https://doi.org/10.1103/PhysRevLett.96.110404} {\bibfield {journal}
{\bibinfo {journal} {Phys. Rev. Lett.}\ }\textbf {\bibinfo {volume} {96}},\
\bibinfo {pages} {110404} (\bibinfo {year} {2006})},\ \Eprint
{https://arxiv.org/abs/hep-th/0510092} {arXiv:hep-th/0510092} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {{Levin}}\ and\ \citenamefont
{{Wen}}(2006)}]{LevinWen2005}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{{Levin}}}\ and\ \bibinfo {author} {\bibfnamefont {X.-G.}\ \bibnamefont
{{Wen}}},\ }\bibfield {title} {\bibinfo {title} {{Detecting Topological
Order in a Ground State Wave Function}},\ }\href
{https://doi.org/10.1103/PhysRevLett.96.110405} {\bibfield {journal}
{\bibinfo {journal} {prl}\ }\textbf {\bibinfo {volume} {96}},\ \bibinfo
{eid} {110405} (\bibinfo {year} {2006})},\ \Eprint
{https://arxiv.org/abs/cond-mat/0510613} {arXiv:cond-mat/0510613
[cond-mat.str-el]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Burnell}}(2018)}]{Burnell2017Review}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {F.~J.}\ \bibnamefont
{{Burnell}}},\ }\bibfield {title} {\bibinfo {title} {{Anyon Condensation and
Its Applications}},\ }\href
{https://doi.org/10.1146/annurev-conmatphys-033117-054154} {\bibfield
{journal} {\bibinfo {journal} {Annual Review of Condensed Matter Physics}\
}\textbf {\bibinfo {volume} {9}},\ \bibinfo {pages} {307} (\bibinfo {year}
{2018})},\ \Eprint {https://arxiv.org/abs/1706.04940} {arXiv:1706.04940
[cond-mat.str-el]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {Witten}(1988)}]{Witten:1988TQFT}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {E.}~\bibnamefont
{Witten}},\ }\bibfield {title} {\bibinfo {title} {{Topological Quantum Field
Theory}},\ }\href {https://doi.org/10.1007/BF01223371} {\bibfield {journal}
{\bibinfo {journal} {Commun. Math. Phys.}\ }\textbf {\bibinfo {volume}
{117}},\ \bibinfo {pages} {353} (\bibinfo {year} {1988})}\BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Fendley}\ \emph {et~al.}(2007)\citenamefont
{Fendley}, \citenamefont {Fisher},\ and\ \citenamefont
{Nayak}}]{Fendley_2007}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {P.}~\bibnamefont
{Fendley}}, \bibinfo {author} {\bibfnamefont {M.~P.~A.}\ \bibnamefont
{Fisher}},\ and\ \bibinfo {author} {\bibfnamefont {C.}~\bibnamefont
{Nayak}},\ }\bibfield {title} {\bibinfo {title} {Topological entanglement
entropy from the holographic partition function},\ }\href
{https://doi.org/10.1007/s10955-006-9275-8} {\bibfield {journal} {\bibinfo
{journal} {Journal of Statistical Physics}\ }\textbf {\bibinfo {volume}
{126}},\ \bibinfo {pages} {1111} (\bibinfo {year} {2007})}\BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Shankar}\ and\ \citenamefont
{Vishwanath}(2011)}]{Shankar_2011}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {R.}~\bibnamefont
{Shankar}}\ and\ \bibinfo {author} {\bibfnamefont {A.}~\bibnamefont
{Vishwanath}},\ }\bibfield {title} {\bibinfo {title} {Equality of bulk wave
functions and edge correlations in some topological superconductors: A
spacetime derivation},\ }\bibfield {journal} {\bibinfo {journal} {Physical
Review Letters}\ }\textbf {\bibinfo {volume} {107}},\ \href
{https://doi.org/10.1103/physrevlett.107.106803}
{10.1103/physrevlett.107.106803} (\bibinfo {year} {2011})\BibitemShut
{NoStop}%
\bibitem [{Note3()}]{Note3}%
\BibitemOpen
\bibinfo {note} {A similar trick was used to study measurement-induced
phenomena in critical systems~\cite {Garratt:2022ycp}.}\BibitemShut {Stop}%
\bibitem [{\citenamefont {{Beigi}}\ \emph {et~al.}(2011)\citenamefont
{{Beigi}}, \citenamefont {{Shor}},\ and\ \citenamefont
{{Whalen}}}]{BeigiShor2011}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {S.}~\bibnamefont
{{Beigi}}}, \bibinfo {author} {\bibfnamefont {P.~W.}\ \bibnamefont
{{Shor}}},\ and\ \bibinfo {author} {\bibfnamefont {D.}~\bibnamefont
{{Whalen}}},\ }\bibfield {title} {\bibinfo {title} {{The Quantum Double
Model with Boundary: Condensations and Symmetries}},\ }\href
{https://doi.org/10.1007/s00220-011-1294-x} {\bibfield {journal} {\bibinfo
{journal} {Communications in Mathematical Physics}\ }\textbf {\bibinfo
{volume} {306}},\ \bibinfo {pages} {663} (\bibinfo {year} {2011})},\ \Eprint
{https://arxiv.org/abs/1006.5479} {arXiv:1006.5479 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {{Kapustin}}\ and\ \citenamefont
{{Saulina}}(2011)}]{Kapustin2011}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {A.}~\bibnamefont
{{Kapustin}}}\ and\ \bibinfo {author} {\bibfnamefont {N.}~\bibnamefont
{{Saulina}}},\ }\bibfield {title} {\bibinfo {title} {{Topological boundary
conditions in abelian Chern-Simons theory}},\ }\href
{https://doi.org/10.1016/j.nuclphysb.2010.12.017} {\bibfield {journal}
{\bibinfo {journal} {Nuclear Physics B}\ }\textbf {\bibinfo {volume}
{845}},\ \bibinfo {pages} {393} (\bibinfo {year} {2011})},\ \Eprint
{https://arxiv.org/abs/1008.0654} {arXiv:1008.0654 [hep-th]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Elitzur}\ \emph {et~al.}(1989)\citenamefont
{Elitzur}, \citenamefont {Moore}, \citenamefont {Schwimmer},\ and\
\citenamefont {Seiberg}}]{Elitzur:1989nr}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {S.}~\bibnamefont
{Elitzur}}, \bibinfo {author} {\bibfnamefont {G.~W.}\ \bibnamefont {Moore}},
\bibinfo {author} {\bibfnamefont {A.}~\bibnamefont {Schwimmer}},\ and\
\bibinfo {author} {\bibfnamefont {N.}~\bibnamefont {Seiberg}},\ }\bibfield
{title} {\bibinfo {title} {{Remarks on the Canonical Quantization of the
Chern-Simons-Witten Theory}},\ }\href
{https://doi.org/10.1016/0550-3213(89)90436-7} {\bibfield {journal}
{\bibinfo {journal} {Nucl. Phys. B}\ }\textbf {\bibinfo {volume} {326}},\
\bibinfo {pages} {108} (\bibinfo {year} {1989})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Wen}(1995)}]{Wen:1995qn}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {X.-G.}\ \bibnamefont
{Wen}},\ }\bibfield {title} {\bibinfo {title} {{Topological orders and edge
excitations in FQH states}},\ }\href
{https://doi.org/10.1080/00018739500101566} {\bibfield {journal} {\bibinfo
{journal} {Adv. Phys.}\ }\textbf {\bibinfo {volume} {44}},\ \bibinfo {pages}
{405} (\bibinfo {year} {1995})},\ \Eprint
{https://arxiv.org/abs/cond-mat/9506066} {arXiv:cond-mat/9506066}
\BibitemShut {NoStop}%
\bibitem [{Note4()}]{Note4}%
\BibitemOpen
\bibinfo {note} {The K-matrix $\protect \mathbb {K}^{(2)}$ flips sign under
the anti-unitary symmetry, however, the Lagrangian $\protect \mathcal {L}_0$
is invariant as the minus sign cancels that from the imaginary
identity.}\BibitemShut {Stop}%
\bibitem [{Note5()}]{Note5}%
\BibitemOpen
\bibinfo {note} {The effective theory for the norm of the EFD $\protect \mbox
{$ \langle \protect \!\langle \rho | \rho \rangle \protect \!\rangle $} =
\protect \operatorname {tr}\rho ^2$ is expected to have a $\protect \mathbb
{Z}_2$ unitary symmetry permuting two copies of density matrix. This
permutation is a combination of the Hermitian conjugation of the EFD and that
of the density matrix and thus is included in $\protect \mathcal
{G}^{(2)}$.}\BibitemShut {Stop}%
\bibitem [{Note6()}]{Note6}%
\BibitemOpen
\bibinfo {note} {The equation is held modulo physical excitations, where
$\Lambda $ is an integer vector. $\protect \bm {1}$ is a vector with each
component being $1$ and of the same dimension as the matrix $K$.}\BibitemShut
{Stop}%
\bibitem [{\citenamefont {Levin}\ and\ \citenamefont
{Gu}(2012)}]{LevinGu:2012}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Levin}}\ and\ \bibinfo {author} {\bibfnamefont {Z.-C.}\ \bibnamefont {Gu}},\
}\bibfield {title} {\bibinfo {title} {{Braiding statistics approach to
symmetry-protected topological phases}},\ }\href
{https://doi.org/10.1103/PhysRevB.86.115109} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. B}\ }\textbf {\bibinfo {volume} {86}},\ \bibinfo
{pages} {115109} (\bibinfo {year} {2012})},\ \Eprint
{https://arxiv.org/abs/1202.3120} {arXiv:1202.3120 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Laughlin}(1983)}]{Laughlin1983}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {R.~B.}\ \bibnamefont
{Laughlin}},\ }\bibfield {title} {\bibinfo {title} {Anomalous quantum hall
effect: An incompressible quantum fluid with fractionally charged
excitations},\ }\href {https://doi.org/10.1103/PhysRevLett.50.1395}
{\bibfield {journal} {\bibinfo {journal} {Phys. Rev. Lett.}\ }\textbf
{\bibinfo {volume} {50}},\ \bibinfo {pages} {1395} (\bibinfo {year}
{1983})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Bravyi}}\ \emph {et~al.}(2018)\citenamefont
{{Bravyi}}, \citenamefont {{Englbrecht}}, \citenamefont {{K{\"o}nig}},\ and\
\citenamefont {{Peard}}}]{Bravyi2018Coherent}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {S.}~\bibnamefont
{{Bravyi}}}, \bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{{Englbrecht}}}, \bibinfo {author} {\bibfnamefont {R.}~\bibnamefont
{{K{\"o}nig}}},\ and\ \bibinfo {author} {\bibfnamefont {N.}~\bibnamefont
{{Peard}}},\ }\bibfield {title} {\bibinfo {title} {{Correcting coherent
errors with surface codes}},\ }\href
{https://doi.org/10.1038/s41534-018-0106-y} {\bibfield {journal} {\bibinfo
{journal} {npj Quantum Information}\ }\textbf {\bibinfo {volume} {4}},\
\bibinfo {eid} {55} (\bibinfo {year} {2018})},\ \Eprint
{https://arxiv.org/abs/1710.02270} {arXiv:1710.02270 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {{Huang}}\ \emph {et~al.}(2019)\citenamefont
{{Huang}}, \citenamefont {{Doherty}},\ and\ \citenamefont
{{Flammia}}}]{Flammia2019QECCoherent}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {E.}~\bibnamefont
{{Huang}}}, \bibinfo {author} {\bibfnamefont {A.~C.}\ \bibnamefont
{{Doherty}}},\ and\ \bibinfo {author} {\bibfnamefont {S.}~\bibnamefont
{{Flammia}}},\ }\bibfield {title} {\bibinfo {title} {{Performance of quantum
error correction with coherent errors}},\ }\href
{https://doi.org/10.1103/PhysRevA.99.022313} {\bibfield {journal} {\bibinfo
{journal} {\pra}\ }\textbf {\bibinfo {volume} {99}},\ \bibinfo {eid} {022313}
(\bibinfo {year} {2019})},\ \Eprint {https://arxiv.org/abs/1805.08227}
{arXiv:1805.08227 [quant-ph]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Iverson}}\ and\ \citenamefont
{{Preskill}}(2020)}]{Preskill2020Coherent}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {J.~K.}\ \bibnamefont
{{Iverson}}}\ and\ \bibinfo {author} {\bibfnamefont {J.}~\bibnamefont
{{Preskill}}},\ }\bibfield {title} {\bibinfo {title} {{Coherence in logical
quantum channels}},\ }\href {https://doi.org/10.1088/1367-2630/ab8e5c}
{\bibfield {journal} {\bibinfo {journal} {New Journal of Physics}\ }\textbf
{\bibinfo {volume} {22}},\ \bibinfo {eid} {073066} (\bibinfo {year}
{2020})},\ \Eprint {https://arxiv.org/abs/1912.04319} {arXiv:1912.04319
[quant-ph]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {Venn}\ \emph {et~al.}(2022)\citenamefont {Venn},
\citenamefont {Behrends},\ and\ \citenamefont {B\'eri}}]{Venn:2022kxy}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {F.}~\bibnamefont
{Venn}}, \bibinfo {author} {\bibfnamefont {J.}~\bibnamefont {Behrends}},\
and\ \bibinfo {author} {\bibfnamefont {B.}~\bibnamefont {B\'eri}},\
}\bibfield {title} {\bibinfo {title} {{Coherent error threshold for surface
codes from Majorana delocalization}},\ }\href@noop {} {\ (\bibinfo {year}
{2022})},\ \Eprint {https://arxiv.org/abs/2211.00655} {arXiv:2211.00655
[quant-ph]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Darmawan}}\ and\ \citenamefont
{{Poulin}}(2017)}]{Poulin2017}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {A.~S.}\ \bibnamefont
{{Darmawan}}}\ and\ \bibinfo {author} {\bibfnamefont {D.}~\bibnamefont
{{Poulin}}},\ }\bibfield {title} {\bibinfo {title} {{Tensor-Network
Simulations of the Surface Code under Realistic Noise}},\ }\href
{https://doi.org/10.1103/PhysRevLett.119.040502} {\bibfield {journal}
{\bibinfo {journal} {\prl}\ }\textbf {\bibinfo {volume} {119}},\ \bibinfo
{eid} {040502} (\bibinfo {year} {2017})},\ \Eprint
{https://arxiv.org/abs/1607.06460} {arXiv:1607.06460 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {{Darmawan}}\ and\ \citenamefont
{{Poulin}}(2018)}]{Poulin2018}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {A.~S.}\ \bibnamefont
{{Darmawan}}}\ and\ \bibinfo {author} {\bibfnamefont {D.}~\bibnamefont
{{Poulin}}},\ }\bibfield {title} {\bibinfo {title} {{Linear-time general
decoding algorithm for the surface code}},\ }\href
{https://doi.org/10.1103/PhysRevE.97.051302} {\bibfield {journal} {\bibinfo
{journal} {\pre}\ }\textbf {\bibinfo {volume} {97}},\ \bibinfo {eid} {051302}
(\bibinfo {year} {2018})},\ \Eprint {https://arxiv.org/abs/1801.01879}
{arXiv:1801.01879 [quant-ph]} \BibitemShut {NoStop}%
\bibitem [{\citenamefont {Fan}\ \emph {et~al.}(pear)\citenamefont {Fan},
\citenamefont {Bao}, \citenamefont {Altman},\ and\ \citenamefont
{Vishwanath}}]{Diagnosticstoappear}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {R.}~\bibnamefont
{Fan}}, \bibinfo {author} {\bibfnamefont {Y.}~\bibnamefont {Bao}}, \bibinfo
{author} {\bibfnamefont {E.}~\bibnamefont {Altman}},\ and\ \bibinfo {author}
{\bibfnamefont {A.}~\bibnamefont {Vishwanath}},\ }\bibfield {title}
{\bibinfo {title} {{Diagnostics of mixed-state topological order and
breakdown of quantum memory}},\ }\href@noop {} {\ (\bibinfo {year} {to
appear})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Haah}(2011)}]{Haah:2011drr}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {J.}~\bibnamefont
{Haah}},\ }\bibfield {title} {\bibinfo {title} {{Local stabilizer codes in
three dimensions without string logical operators}},\ }\href
{https://doi.org/10.1103/physreva.83.042330} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. A}\ }\textbf {\bibinfo {volume} {83}},\ \bibinfo
{pages} {042330} (\bibinfo {year} {2011})},\ \Eprint
{https://arxiv.org/abs/1101.1962} {arXiv:1101.1962 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Vijay}\ \emph {et~al.}(2016)\citenamefont {Vijay},
\citenamefont {Haah},\ and\ \citenamefont {Fu}}]{Vijay:2016phm}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {S.}~\bibnamefont
{Vijay}}, \bibinfo {author} {\bibfnamefont {J.}~\bibnamefont {Haah}},\ and\
\bibinfo {author} {\bibfnamefont {L.}~\bibnamefont {Fu}},\ }\bibfield
{title} {\bibinfo {title} {{Fracton Topological Order, Generalized Lattice
Gauge Theory and Duality}},\ }\href
{https://doi.org/10.1103/PhysRevB.94.235157} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. B}\ }\textbf {\bibinfo {volume} {94}},\ \bibinfo
{pages} {235157} (\bibinfo {year} {2016})},\ \Eprint
{https://arxiv.org/abs/1603.04442} {arXiv:1603.04442 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Breuckmann}\ and\ \citenamefont
{Eberhardt}(2021)}]{Breuckmann:2021yvk}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {N.~P.}\ \bibnamefont
{Breuckmann}}\ and\ \bibinfo {author} {\bibfnamefont {J.~N.}\ \bibnamefont
{Eberhardt}},\ }\bibfield {title} {\bibinfo {title} {{Quantum Low-Density
Parity-Check Codes}},\ }\href {https://doi.org/10.1103/PRXQuantum.2.040101}
{\bibfield {journal} {\bibinfo {journal} {PRX Quantum}\ }\textbf {\bibinfo
{volume} {2}},\ \bibinfo {pages} {040101} (\bibinfo {year} {2021})},\ \Eprint
{https://arxiv.org/abs/2103.06309} {arXiv:2103.06309 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Ma}\ and\ \citenamefont {Wang}(2022)}]{Ma:2022pvq}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {R.}~\bibnamefont
{Ma}}\ and\ \bibinfo {author} {\bibfnamefont {C.}~\bibnamefont {Wang}},\
}\bibfield {title} {\bibinfo {title} {{Average Symmetry-Protected
Topological Phases}},\ }\href@noop {} {\ (\bibinfo {year} {2022})},\ \Eprint
{https://arxiv.org/abs/2209.02723} {arXiv:2209.02723 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Lee}\ \emph {et~al.}()\citenamefont {Lee},
\citenamefont {Jian},\ and\ \citenamefont {Xu}}]{CenkeToAppear}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {J.~Y.}\ \bibnamefont
{Lee}}, \bibinfo {author} {\bibfnamefont {C.-M.}\ \bibnamefont {Jian}},\ and\
\bibinfo {author} {\bibfnamefont {C.}~\bibnamefont {Xu}},\ }\bibfield
{title} {\bibinfo {title} {{Quantum criticality under decoherence or
weak-measurement}},\ }\href@noop {} {\bibinfo {journal} {to appear}\
}\BibitemShut {NoStop}%
\end{thebibliography}%
%merlin.mbs apsrev4-1.bst 2010-07-25 4.21a (PWD, AO, DPC) hacked
%Control: key (0)
%Control: author (8) initials jnrlst
%Control: editor formatted (1) identically to author
%Control: production of article title (-1) disabled
%Control: page (0) single
%Control: year (1) truncated
%Control: production of eprint (0) enabled
\begin{thebibliography}{14}%
\makeatletter
\providecommand \@ifxundefined [1]{%
\@ifx{#1\undefined}
}%
\providecommand \@ifnum [1]{%
\ifnum #1\expandafter \@firstoftwo
\else \expandafter \@secondoftwo
\fi
}%
\providecommand \@ifx [1]{%
\ifx #1\expandafter \@firstoftwo
\else \expandafter \@secondoftwo
\fi
}%
\providecommand \natexlab [1]{#1}%
\providecommand \enquote [1]{``#1''}%
\providecommand \bibnamefont [1]{#1}%
\providecommand \bibfnamefont [1]{#1}%
\providecommand \citenamefont [1]{#1}%
\providecommand \href@noop [0]{\@secondoftwo}%
\providecommand \href [0]{\begingroup \@sanitize@url \@href}%
\providecommand \@href[1]{\@@startlink{#1}\@@href}%
\providecommand \@@href[1]{\endgroup#1\@@endlink}%
\providecommand \@sanitize@url [0]{\catcode `\\12\catcode `\$12\catcode
`\&12\catcode `\#12\catcode `\^12\catcode `\_12\catcode `\%12\relax}%
\providecommand \@@startlink[1]{}%
\providecommand \@@endlink[0]{}%
\providecommand \url [0]{\begingroup\@sanitize@url \@url }%
\providecommand \@url [1]{\endgroup\@href {#1}{\urlprefix }}%
\providecommand \urlprefix [0]{URL }%
\providecommand \Eprint [0]{\href }%
\providecommand \doibase [0]{http://dx.doi.org/}%
\providecommand \selectlanguage [0]{\@gobble}%
\providecommand \bibinfo [0]{\@secondoftwo}%
\providecommand \bibfield [0]{\@secondoftwo}%
\providecommand \translation [1]{[#1]}%
\providecommand \BibitemOpen [0]{}%
\providecommand \bibitemStop [0]{}%
\providecommand \bibitemNoStop [0]{.\EOS\space}%
\providecommand \EOS [0]{\spacefactor3000\relax}%
\providecommand \BibitemShut [1]{\csname bibitem#1\endcsname}%
\let\auto@bib@innerbib\@empty
%</preamble>
\bibitem [{\citenamefont {{Castelnovo}}\ and\ \citenamefont
{{Chamon}}(2008)}]{CastelnovoPRB2008}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {C.}~\bibnamefont
{{Castelnovo}}}\ and\ \bibinfo {author} {\bibfnamefont {C.}~\bibnamefont
{{Chamon}}},\ }\href {\doibase 10.1103/PhysRevB.77.054433} {\bibfield
{journal} {\bibinfo {journal} {\prb}\ }\textbf {\bibinfo {volume} {77}},\
\bibinfo {eid} {054433} (\bibinfo {year} {2008})},\ \Eprint
{http://arxiv.org/abs/0707.2084} {arXiv:0707.2084 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{Note1()}]{Note1}%
\BibitemOpen
\bibinfo {note} {We thank Meng Cheng for pointing this out to
us.}\BibitemShut {Stop}%
\bibitem [{\citenamefont {Read}(1990)}]{Read:1990zz}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {N.}~\bibnamefont
{Read}},\ }\href {\doibase 10.1103/PhysRevLett.65.1502} {\bibfield {journal}
{\bibinfo {journal} {Phys. Rev. Lett.}\ }\textbf {\bibinfo {volume} {65}},\
\bibinfo {pages} {1502} (\bibinfo {year} {1990})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Wen}\ and\ \citenamefont {Zee}(1992)}]{Wen:1992uk}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {X.~G.}\ \bibnamefont
{Wen}}\ and\ \bibinfo {author} {\bibfnamefont {A.}~\bibnamefont {Zee}},\
}\href {\doibase 10.1103/PhysRevB.46.2290} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. B}\ }\textbf {\bibinfo {volume} {46}},\ \bibinfo
{pages} {2290} (\bibinfo {year} {1992})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Frohlich}\ and\ \citenamefont
{Zee}(1991)}]{Frohlich:1991wb}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {J.}~\bibnamefont
{Frohlich}}\ and\ \bibinfo {author} {\bibfnamefont {A.}~\bibnamefont {Zee}},\
}\href {\doibase 10.1016/0550-3213(91)90275-3} {\bibfield {journal}
{\bibinfo {journal} {Nucl. Phys. B}\ }\textbf {\bibinfo {volume} {364}},\
\bibinfo {pages} {517} (\bibinfo {year} {1991})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Wen}(1995)}]{Wen:1995qn}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {X.-G.}\ \bibnamefont
{Wen}},\ }\href {\doibase 10.1080/00018739500101566} {\bibfield {journal}
{\bibinfo {journal} {Adv. Phys.}\ }\textbf {\bibinfo {volume} {44}},\
\bibinfo {pages} {405} (\bibinfo {year} {1995})},\ \Eprint
{http://arxiv.org/abs/cond-mat/9506066} {arXiv:cond-mat/9506066} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {Levin}(2013)}]{Levin:2013gaa}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Levin}},\ }\href {\doibase 10.1103/PhysRevX.3.021009} {\bibfield {journal}
{\bibinfo {journal} {Phys. Rev. X}\ }\textbf {\bibinfo {volume} {3}},\
\bibinfo {pages} {021009} (\bibinfo {year} {2013})},\ \Eprint
{http://arxiv.org/abs/1301.7355} {arXiv:1301.7355 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Barkeshli}\ \emph {et~al.}(2013)\citenamefont
{Barkeshli}, \citenamefont {Jian},\ and\ \citenamefont
{Qi}}]{Barkeshli:2013}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Barkeshli}}, \bibinfo {author} {\bibfnamefont {C.-M.}\ \bibnamefont {Jian}},
\ and\ \bibinfo {author} {\bibfnamefont {X.-L.}\ \bibnamefont {Qi}},\ }\href
{\doibase 10.1103/PhysRevB.88.235103} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. B}\ }\textbf {\bibinfo {volume} {88}},\ \bibinfo
{pages} {235103} (\bibinfo {year} {2013})},\ \Eprint
{http://arxiv.org/abs/1305.7203} {arXiv:1305.7203 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{Note2()}]{Note2}%
\BibitemOpen
\bibinfo {note} {Here and in the following, Table~I refers to the table in
the main text.}\BibitemShut {Stop}%
\bibitem [{\citenamefont {Lichtman}\ \emph {et~al.}(2021)\citenamefont
{Lichtman}, \citenamefont {Thorngren}, \citenamefont {Lindner}, \citenamefont
{Stern},\ and\ \citenamefont {Berg}}]{Lichtman:2020nuw}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {T.}~\bibnamefont
{Lichtman}}, \bibinfo {author} {\bibfnamefont {R.}~\bibnamefont {Thorngren}},
\bibinfo {author} {\bibfnamefont {N.~H.}\ \bibnamefont {Lindner}}, \bibinfo
{author} {\bibfnamefont {A.}~\bibnamefont {Stern}}, \ and\ \bibinfo {author}
{\bibfnamefont {E.}~\bibnamefont {Berg}},\ }\href {\doibase
10.1103/PhysRevB.104.075141} {\bibfield {journal} {\bibinfo {journal}
{Phys. Rev. B}\ }\textbf {\bibinfo {volume} {104}},\ \bibinfo {pages}
{075141} (\bibinfo {year} {2021})},\ \Eprint
{http://arxiv.org/abs/2003.04328} {arXiv:2003.04328 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Levin}\ and\ \citenamefont
{Gu}(2012)}]{LevinGu:2012}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {M.}~\bibnamefont
{Levin}}\ and\ \bibinfo {author} {\bibfnamefont {Z.-C.}\ \bibnamefont {Gu}},\
}\href {\doibase 10.1103/PhysRevB.86.115109} {\bibfield {journal} {\bibinfo
{journal} {Phys. Rev. B}\ }\textbf {\bibinfo {volume} {86}},\ \bibinfo
{pages} {115109} (\bibinfo {year} {2012})},\ \Eprint
{http://arxiv.org/abs/1202.3120} {arXiv:1202.3120 [cond-mat.str-el]}
\BibitemShut {NoStop}%
\bibitem [{\citenamefont {Laughlin}(1983)}]{Laughlin1983}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {R.~B.}\ \bibnamefont
{Laughlin}},\ }\href {\doibase 10.1103/PhysRevLett.50.1395} {\bibfield
{journal} {\bibinfo {journal} {Phys. Rev. Lett.}\ }\textbf {\bibinfo
{volume} {50}},\ \bibinfo {pages} {1395} (\bibinfo {year}
{1983})}\BibitemShut {NoStop}%
\bibitem [{\citenamefont {{Beigi}}\ \emph {et~al.}(2011)\citenamefont
{{Beigi}}, \citenamefont {{Shor}},\ and\ \citenamefont
{{Whalen}}}]{BeigiShor2011}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {S.}~\bibnamefont
{{Beigi}}}, \bibinfo {author} {\bibfnamefont {P.~W.}\ \bibnamefont {{Shor}}},
\ and\ \bibinfo {author} {\bibfnamefont {D.}~\bibnamefont {{Whalen}}},\
}\href {\doibase 10.1007/s00220-011-1294-x} {\bibfield {journal} {\bibinfo
{journal} {Communications in Mathematical Physics}\ }\textbf {\bibinfo
{volume} {306}},\ \bibinfo {pages} {663} (\bibinfo {year} {2011})},\ \Eprint
{http://arxiv.org/abs/1006.5479} {arXiv:1006.5479 [quant-ph]} \BibitemShut
{NoStop}%
\bibitem [{\citenamefont {{Kapustin}}\ and\ \citenamefont
{{Saulina}}(2011)}]{Kapustin2011}%
\BibitemOpen
\bibfield {author} {\bibinfo {author} {\bibfnamefont {A.}~\bibnamefont
{{Kapustin}}}\ and\ \bibinfo {author} {\bibfnamefont {N.}~\bibnamefont
{{Saulina}}},\ }\href {\doibase 10.1016/j.nuclphysb.2010.12.017} {\bibfield
{journal} {\bibinfo {journal} {Nuclear Physics B}\ }\textbf {\bibinfo
{volume} {845}},\ \bibinfo {pages} {393} (\bibinfo {year} {2011})},\ \Eprint
{http://arxiv.org/abs/1008.0654} {arXiv:1008.0654 [hep-th]} \BibitemShut
{NoStop}%
\end{thebibliography}%