Statistical Analysis of Quantum State Learning Process in Quantum Neural Networks
Authors: Hao-Kai Zhang, Chenghong Zhu, Mingrui Jing, Xin Wang
NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive numerical simulations are performed to validate our theoretical results. Our findings place generic limits on good initial guesses and adaptive methods for improving the learnability and scalability of QNNs, and deepen the understanding of prior information s role in QNNs. We conduct extensive numerical experiments to verify our theoretical findings. |
| Researcher Affiliation | Collaboration | Hao-kai Zhang1,3, Chenghong Zhu2,3, Mingrui Jing2,3, Xin Wang2,3 1 Institute for Advanced Study, Tsinghua University, Beijing 100084, China 2 Thrust of Artificial Intelligence, Information Hub, Hong Kong University of Science and Technology (Guangzhou), China 3 Institute for Quantum Computing, Baidu Research, Beijing, China |
| Pseudocode | No | The paper does not contain any pseudocode or algorithm blocks. It provides mathematical derivations, theorems, and circuit diagrams, but no structured algorithmic steps. |
| Open Source Code | Yes | The codes for numerical experiments can be found in [75]. [75] Numerical Experiments of this work. https://github.com/chenghongz/lim_learning_state. |
| Open Datasets | No | The paper describes generating data for experiments by sampling 'target states from the ensemble T' but does not use a publicly available or open dataset with a specific name, citation, or direct access link in the conventional sense. |
| Dataset Splits | No | The paper does not explicitly provide training, validation, or test dataset splits. It describes sampling target states for simulations and comparing theoretical bounds with loss curves but does not specify data partitioning percentages or counts. |
| Hardware Specification | No | The paper does not explicitly describe any specific hardware components (e.g., GPU/CPU models, memory, cloud instances) used to run the experiments. It only mentions using 'Paddle Quantum [73] and Tensorcircuit [74]' which are software platforms. |
| Software Dependencies | No | The paper mentions using 'Paddle Quantum [73] and Tensorcircuit [74]' platforms for numerical experiments, but it does not specify version numbers for these software components or any other libraries. |
| Experiment Setup | Yes | We create 9 ALT circuits for qubit counts of 2, 6, 10 and circuit depth of 1, 3, 5 with randomly initialized parameters, denoted as θ . For each circuit, we sample 10 target states from the ensemble T with p = 0.2 and then generate 10 corresponding loss curves using the Adam optimizer with a learning rate 0.01. |