Post-variational classical quantum transfer learning for binary classification
Abstract We address the limitations of variational quantum circuits (VQCs) in hybrid classical-quantum transfer learning by introducing post-variational strategies, which reduce training overhead and mitigate optimization issues. Our approach Post Variational Classical Quantum Transfer Learning (PVC...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-08887-2 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849238842856439808 |
|---|---|
| author | Kavitha Yogaraj Brian Quanz Tarun Vikas Arijit Mondal Samrat Mondal |
| author_facet | Kavitha Yogaraj Brian Quanz Tarun Vikas Arijit Mondal Samrat Mondal |
| author_sort | Kavitha Yogaraj |
| collection | DOAJ |
| description | Abstract We address the limitations of variational quantum circuits (VQCs) in hybrid classical-quantum transfer learning by introducing post-variational strategies, which reduce training overhead and mitigate optimization issues. Our approach Post Variational Classical Quantum Transfer Learning (PVCQTL) includes three designs: (1) modified observable construction, (2) a hybrid approach, and (3) a variational-post-variational combination. We evaluate these on pre-trained models (VGG19, ResNet50, ResNet18, MobileNet) for 4 and 8 qubits, with ResNet50 performing best in deepfake detection. Compared to classical models (MLP, ResNet50) and quantum baselines hybrid quantum classical neural network (HQCNN), classical-quantum transfer learning (CQTL). PVCQTL consistently achieves better accuracy. The modified observable variant reaches 85% accuracy for Deepfake dataset with lower computational cost. To evaluate generalizability, we tested PVCQTL on three additional binary classification datasets, observing improved accuracy on each. We conducted ablation studies to assess the effects of architectural choices on quantum component variations, including the choice of quantum gates, use of fixed ansatz circuits, and observable measurements. Robustness to input noise and sensitivity of the PVCQTL models were examined through ablation studies on learning rate, batch size, and number of qubits. These results demonstrate that PVCQTL offers a measurable improvement over traditional hybrid classical-quantum approaches. |
| format | Article |
| id | doaj-art-4a638f351c6f4b77ae55b37ccf4870fc |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-4a638f351c6f4b77ae55b37ccf4870fc2025-08-20T04:01:24ZengNature PortfolioScientific Reports2045-23222025-07-0115112210.1038/s41598-025-08887-2Post-variational classical quantum transfer learning for binary classificationKavitha Yogaraj0Brian Quanz1Tarun Vikas2Arijit Mondal3Samrat Mondal4IBM Quantum, IBM ResearchIBM Quantum, IBM ResearchDepartment of Computer Science and Engineering, Indian Institute of TechnologyDepartment of Computer Science and Engineering, Indian Institute of TechnologyDepartment of Computer Science and Engineering, Indian Institute of TechnologyAbstract We address the limitations of variational quantum circuits (VQCs) in hybrid classical-quantum transfer learning by introducing post-variational strategies, which reduce training overhead and mitigate optimization issues. Our approach Post Variational Classical Quantum Transfer Learning (PVCQTL) includes three designs: (1) modified observable construction, (2) a hybrid approach, and (3) a variational-post-variational combination. We evaluate these on pre-trained models (VGG19, ResNet50, ResNet18, MobileNet) for 4 and 8 qubits, with ResNet50 performing best in deepfake detection. Compared to classical models (MLP, ResNet50) and quantum baselines hybrid quantum classical neural network (HQCNN), classical-quantum transfer learning (CQTL). PVCQTL consistently achieves better accuracy. The modified observable variant reaches 85% accuracy for Deepfake dataset with lower computational cost. To evaluate generalizability, we tested PVCQTL on three additional binary classification datasets, observing improved accuracy on each. We conducted ablation studies to assess the effects of architectural choices on quantum component variations, including the choice of quantum gates, use of fixed ansatz circuits, and observable measurements. Robustness to input noise and sensitivity of the PVCQTL models were examined through ablation studies on learning rate, batch size, and number of qubits. These results demonstrate that PVCQTL offers a measurable improvement over traditional hybrid classical-quantum approaches.https://doi.org/10.1038/s41598-025-08887-2Post variationalQuantum transfer learningQuantum binary classification |
| spellingShingle | Kavitha Yogaraj Brian Quanz Tarun Vikas Arijit Mondal Samrat Mondal Post-variational classical quantum transfer learning for binary classification Scientific Reports Post variational Quantum transfer learning Quantum binary classification |
| title | Post-variational classical quantum transfer learning for binary classification |
| title_full | Post-variational classical quantum transfer learning for binary classification |
| title_fullStr | Post-variational classical quantum transfer learning for binary classification |
| title_full_unstemmed | Post-variational classical quantum transfer learning for binary classification |
| title_short | Post-variational classical quantum transfer learning for binary classification |
| title_sort | post variational classical quantum transfer learning for binary classification |
| topic | Post variational Quantum transfer learning Quantum binary classification |
| url | https://doi.org/10.1038/s41598-025-08887-2 |
| work_keys_str_mv | AT kavithayogaraj postvariationalclassicalquantumtransferlearningforbinaryclassification AT brianquanz postvariationalclassicalquantumtransferlearningforbinaryclassification AT tarunvikas postvariationalclassicalquantumtransferlearningforbinaryclassification AT arijitmondal postvariationalclassicalquantumtransferlearningforbinaryclassification AT samratmondal postvariationalclassicalquantumtransferlearningforbinaryclassification |