Dissipative phase transitions and passive error correction
Yu-Jie Liu, Simon Lieu

TL;DR
This paper classifies passive error correction methods in local Lindblad models, linking them to phases of matter, and demonstrates how certain models exhibit robust steady state degeneracy for protecting classical and quantum information.
Contribution
It introduces a definition of dissipative phases based on steady state degeneracy and analyzes models like the 2D Ising and toric codes to connect phase properties with error correction.
Findings
2D Ising model has robust classical degeneracy at low temperature
4D toric code exhibits robust quantum degeneracy
Perturbations do not alter qualitative features of steady state degeneracy
Abstract
We classify different ways to passively protect classical and quantum information, i.e. we do not allow for syndrome measurements, in the context of local Lindblad models for spin systems. Within this family of models, we suggest that passive error correction is associated with nontrivial phases of matter and propose a definition for dissipative phases based on robust steady state degeneracy of a Lindbladian in the thermodynamic limit. We study three thermalizing models in this context: the 2D Ising model, the 2D toric code, and the 4D toric code. In the low-temperature phase, the 2D Ising model hosts a robust classical steady state degeneracy while the 4D toric code hosts a robust quantum steady state degeneracy. We perturb the models with terms that violate detailed balance and observe that qualitative features remain unchanged, suggesting that symmetry breaking in a…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsQuantum many-body systems · Quantum Computing Algorithms and Architecture · Quantum and electron transport phenomena
