arXiv：cs.LG· Devroop Kar, Daniel Krutz, Travis Desell·· 3 小时前AI 评分27
EXAQC 进化框架实现图像分类的混合量子-经典架构自动搜索
Evolving Hybrid Quantum-Classical Architectures for Image Classification
AI 导读
EXAQC 进化框架被扩展至图像分类任务，自动搜索参数化量子电路作为中间处理模块，在 MNIST、Fashion-MNIST 和 CIFAR-10 上分别达到 98.42%、90.62% 和 85.47% 的准确率。
正文
Abstract:Hybrid quantum classical neural networks integrate parameterized quantum circuits (PQCs) with established deep learning architectures, but their performance depends strongly on the choice of quantum circuit architecture, a choice that remains largely manual. Most existing approaches rely on hand-designed or fixed circuit ansätze, requiring circuit structure, gate composition, and qubit connectivity to be specified in advance with no guarantee that they suit the task. This limitation is especially acute in image classification, where quantum circuits must transform features extracted by classical networks while remaining compact enough for practical training, requirements that generic, task-agnostic ansätze are unlikely to satisfy simultaneously. We extend EXAQC, an evolutionary framework for automated quantum circuit discovery, to image classification. EXAQC evolves PQCs as intermediate processing modules while retaining classical feature-extraction and prediction layers. On MNIST, Fashion-MNIST, and CIFAR-10, EXAQC achieves 98.42%, 90.62%, and 85.47% accuracy, respectively, while using comparable gate counts to other quantum architecture-search methods. Against classical networks, evolved hybrid models maintain comparable accuracy with substantially fewer trainable parameters, reaching 85.68% on CIFAR-10 with over 25$\times$ fewer parameters than a 10-layer CNN. Encoding choice also matters: rotation-based encodings (RX, RY, U3) outperform amplitude encoding by 22-25 points on CIFAR-10. These results demonstrate that automated circuit discovery yields compact quantum modules that can replace larger classical components in vision architectures while retaining competitive accuracy.
| Comments: | Under Review at The Fifteenth International Conference on Learning Representations 2027 |
| Subjects: | Neural and Evolutionary Computing (cs.NE); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Quantum Physics (quant-ph) |
| Cite as: | arXiv:2610.03220 [cs.NE] |
| (or arXiv:2610.03220v1 [cs.NE] for this version) | |
| https://doi.org/10.48550/arXiv.2610.03220 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Devroop Kar [view email]
[v1]
Fri, 2 Oct 2026 12:37:10 UTC (3,616 KB)
来源：arXiv：cs.LG · arxiv.org