Quantum Error Correction: From Memory to Logic
Year: 2026
Type: dissertation
Abstract: Quantum computers possess computational power distinct from classical devices, using phase coherence and entanglement. This coherence is fragile and susceptible to environmental noise. Protecting quantum information is a central goal of quantum error correction and fault tolerance, which are essential for scaling quantum devices to utility-scale computation. Such protection, however, often requires a space-time overhead formidable for current experimental platforms. A key question is the minimum space-time cost of protecting a quantum information-processing task and how to attain it. This thesis investigates these questions from quantum memory to logical gates. For quantum memory, we consider preserving information while it idles under noise. We first develop the near-optimal fidelity, an optimization-free general framework that bounds the optimal performance of a given encoding and noise channel and provides an achievable recovery benchmark. We apply it to Gottesman-Kitaev-Preskill (GKP) codes, which encode information in quantum harmonic oscillators through lattice geometry. We show that finite-size GKP memories can substantially outperform existing recovery schemes. In the asymptotic system size and energy limit, GKP families achieve the capacities of pure-loss and quantum-limited amplification channels, attaining the minimum encoding redundancy permitted by these channels. For logical gates, we study quantum low-density parity-check codes. We introduce high-rate surgery, which performs parallel, addressable logical Pauli-product measurements with low space-time overhead. Our analytical constructions and complementary numerical benchmarks indicate that logical processing can retain nearly constant space-time overhead relative to quantum memory. These results provide efficient building blocks for utility-scale quantum devices.
Source: University of Chicago
DOI: 10.6082/fen77-2qt79
Author: Zheng, Guo
Institution: University of Chicago
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7225095222
