aalto1 untyped-item.component.html
Accelerating searching for quantum error correction codes with reinforcement learning
Loading...
URL
Journal Title
Journal ISSN
Volume Title
School of Science |
Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Authors
Date
Department
Major/Subject
Mcode
Language
en
Pages
50
Series
Abstract
Quantum Error Correction (QEC) codes are a necessary component of utility-scale quantum computers, and the discovery of new codes with desirable properties is one of the most promising avenues for reducing the estimated resources required to perform useful quantum computations. Most current methods for code discovery rely on analytical techniques and construct codes from algebraic or geometric structures, but there also exist some techniques for discovering codes via computational search.
In this work we focus on one of the most promising of these methods where a reinforcement learning agent is trained to construct encoding circuits and evaluates the performance of the corresponding code. We seek to improve upon this method by reducing the computational resources required to discover codes of given parameters, primarily by utilizing a probabilistic approximation of the reward function which relies on randomly sampling error operators according to some specified distribution.
We investigate the performance of our modified framework and compare the discovered codes and required computational resources to the original work, concluding that probabilistic methods such as our sampling technique are effective strategies for reducing the computational cost of the evaluation component of code searching without excessive penalties to performance.