Comparing neural network based decoders for the surface code
arXiv:1811.12456 · doi:10.1109/TC.2019.2948612
Abstract
Matching algorithms can be used for identifying errors in quantum systems, being the most famous the Blossom algorithm. Recent works have shown that small distance quantum error correction codes can be efficiently decoded by employing machine learning techniques based on neural networks (NN). Various NN-based decoders have been proposed to enhance the decoding performance and the decoding time. Their implementation differs in how the decoding is performed, at logical or physical level, as well as in several neural network related parameters. In this work, we implement and compare two NN-based decoders, a low level decoder and a high level decoder, and study how different NN parameters affect their decoding performance and execution time. Crucial parameters such as the size of the training dataset, the structure and the type of the neural network, and the learning rate used during training are discussed. After performing this comparison, we conclude that the high level decoder based on a Recurrent NN shows a better balance between decoding performance and execution time and it is much easier to train. We then test its decoding performance for different code distances, probability datasets and under the depolarizing and circuit error models.
arXiv comment18 pages, 13 figures
References in corpus (27)
- Fault-tolerant quantum computation by anyons
- Surface codes: Towards practical large-scale quantum computation
- Topological quantum memory
- Fault-tolerant quantum computation with high threshold in two dimensions
- Surface code quantum computing by lattice surgery
- High threshold universal quantum computation on the surface code
- Quantum computing with nearest neighbor interactions and error rates over 1%
- Fast Decoders for Topological Quantum Codes
- Efficient Algorithms for Maximum Likelihood Decoding in the Surface Code
- Towards practical classical processing for the surface code
- Scalable quantum circuit and control for a superconducting surface code
- A Neural Decoder for Topological Codes
- Optimal Resources for Topological 2D Stabilizer Codes: Comparative Study
- Strong Resilience of Topological Codes to Depolarization
- Machine-learning-assisted correction of correlated qubit errors in a topological code
- Decoding Small Surface Codes with Feedforward Neural Networks
- Quantum Measurements and Gates by Code Deformation
- Deep Neural Network Probabilistic Decoder for Stabilizer Codes
- High threshold error correction for the surface code
- Neural Belief-Propagation Decoders for Quantum Error-Correcting Codes
- Density-matrix simulation of small surface codes under current and projected experimental noise
- An Experimental Microarchitecture for a Superconducting Quantum Processor
- Deep neural decoders for near term fault-tolerant experiments
- Advantages of versatile neural-network decoding for topological codes
- Clifford Gates by Code Deformation
- Scalable Neural Network Decoders for Higher Dimensional Quantum Codes
- Topological code Autotune
Cited by in corpus (25)
- Optimizing Quantum Error Correction Codes with Reinforcement Learning
- Learning to Decode the Surface Code with a Recurrent, Transformer-Based Neural Network
- Neural-Network Decoders for Quantum Error Correction using Surface Codes:A Space Exploration of the Hardware Cost-Performance Trade-Offs
- Reinforcement Learning Decoders for Fault-Tolerant Quantum Computation
- Decoding algorithms for surface codes
- A scalable and fast artificial neural network syndrome decoder for surface codes
- Recent Advances in Deep Learning for Channel Coding: A Survey
- Techniques for combining fast local decoders with global decoders under circuit-level noise
- Neural ensemble decoding for topological quantum error-correcting codes
- Determination of the Semion Code Threshold using Neural Decoders
- Convolutional neural network based decoders for surface codes
- Neural network decoder for near-term surface-code experiments
- Artificial Neural Network Syndrome Decoding on IBM Quantum Processors
- Surface Code Design for Asymmetric Error Channels
- A NEAT Quantum Error Decoder
- Introduction to Quantum Error Correction with Stabilizer Codes
- A blindness property of the Min-Sum decoding for the toric code
- Enhancing Quantum Memory Lifetime with Measurement-Free Local Error Correction and Reinforcement Learning
- QECOOL: On-Line Quantum Error Correction with a Superconducting Decoder for Surface Code
- Superconducting qubits in the millions: the potential and limitations of modularity
- Efficient Syndrome Decoder for Heavy Hexagonal QECC via Machine Learning
- Self-attention U-Net decoder for toric codes
- Subsystem many-hypercube codes: High-rate concatenated codes with low-weight syndrome measurements
- On the interpretability of neural network decoders
- Challenges and opportunities in the supervised learning of quantum circuit outputs