Simultaneous discovery of quantum error correction codes and encoders with a noise-aware reinforcement learning agent
Problem
Discovering quantum error correction codes and encoding circuits tailored to specific qubit hardware is critical and hard to do manually; prior reinforcement learning approaches were severely limited in scalability.
Why it matters
Quantum error correction is essential for building reliable, useful quantum computers.
What the AI did
Trained a reinforcement learning agent that automatically discovers both QEC codes and their encoding circuits from scratch for a given gate set, connectivity and error model, using a Knill-Laflamme-based reward and a vectorized Clifford simulator; also introduced a noise-aware meta-agent concept.
Results
Effective up to 25 physical qubits and distance-5 codes; roadmap stated toward 100 qubits and distance-10 codes.
Impact
Significantly expands the scale of automated QEC code and encoder discovery, with a stated path to larger systems.
Record
| Venue | npj Quantum Information |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Jan Ollé, Remmy A. M. Zen, Matteo Ballerini Puviani, Florian Marquardt |
| Citations | 23 (OpenAlex, at import time) |
| References | 48 |
| Model or method | Reinforcement learning agent (noise-aware meta-agent) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41534-024-00920-y |
| Open access | gold |
| Keywords | quantum error correction, reinforcement learning, code discovery, encoding circuits, noise-aware agent |
Importance rating
significant — thousands to millions affected
Quantum error correction is called a critical problem for experimental quantum computing; work significantly expands automated QEC code discovery scale, a foundational field.
More in Physics & Astronomy
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Magnetic control of tokamak plasmas through deep reinforcement learning | 2022 | 739 |
7
|
| 2 | Solving the quantum many-body problem with artificial neural networks | 2017 | 2,205 |
7
|
| 3 | Learning high-accuracy error decoding for quantum processors | 2024 | 90 |
7
|
| 4 | Optimizing Quantum Error Correction Codes with Reinforcement Learning | 2019 | 158 |
7
|
| 5 | Machine-learning-assisted correction of correlated qubit errors in a topological code | 2018 | 141 |
7
|