AI-Powered Algorithm-Centric Quantum Processor Topology Design
Authors: Tian Li, Xiao-Yue Xu, Chen Ding, Tian-Ci Tian, Wei-You Liao, Shuo Zhang, He-Liang Huang
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments demonstrate that we have achieved notable enhancements in circuit performance, with a minimum of 20% reduction in circuit depth in 60% of the cases examined, and a maximum enhancement of up to 46%. |
| Researcher Affiliation | Academia | 1Henan Key Laboratory of Quantum infommation and Cryptography Zhengzhou, Henan, 450000, China. EMAIL |
| Pseudocode | Yes | Details regarding the hyper-parameters and the pseudo-code for the Reward-Replay Proximal Policy Optimization (RR-PPO) are provided in the Appendix. |
| Open Source Code | Yes | Code https://github.com/qclab-quantum/Qtailor |
| Open Datasets | Yes | Benchmarks and Compared Method . Building upon the foundation established by previous research, we have selected MQT Bench (Quetschlich, Burgholzer, and Wille 2023) as our benchmark toolkit. |
| Dataset Splits | No | The paper uses the MQT Bench benchmark toolkit and evaluates circuit depth multiple times to average results. While it mentions different circuit sizes for evaluation, it does not provide specific train/test/validation splits for the circuits used to train the RL agent itself. The RL agent learns to dynamically tailor topologies for individual circuits, which might not involve a traditional fixed dataset split for its learning process in the typical ML sense. |
| Hardware Specification | No | We gratefully acknowledge Hefei Advanced Computing Center for hardware support with the numerical experiments. However, no specific details about the hardware (e.g., GPU models, CPU types, memory) are provided. |
| Software Dependencies | No | Our experiments were conducted using Qiskit (Contributors and IBM 2024) as the backend framework... To monitor the training process and evaluate efficiency, we utilized Tensor Board (Abadi et al. 2015). The paper mentions Qiskit and Tensor Board with citation years but does not provide specific version numbers for these software dependencies. |
| Experiment Setup | Yes | The detailed hyperparameters are given in Appendix. |