Kernel methods serve as powerful tools to capture nonlinear patterns behind data in machine learning. Quantum kernels, by mapping data into exponentially large Hilbert spaces, offer the potential to solve problems intractable for classical models. However, existing studies encounter performance bottlenecks when processing complex data with localized structural patterns, stemming from the limitation in single-view feature representation and the exclusive reliance on global data structure. In this paper, we propose quantum multiview kernel learning with local information, called L-QMVKL. Specifically, based on the multiple kernel learning, a representative method for multiview data processing, we construct the quantum multiple kernel that combines view-specific quantum kernels to effectively fuse cross-view information. Further leveraging local information to capture intrinsic structural information, we design a sequential training strategy for the quantum circuit parameters and combination weights with the use of the hybrid global-local kernel alignment. We evaluate the effectiveness of L-QMVKL through comprehensive numerical simulations on the Mfeat dataset, demonstrating significant accuracy improvements achieved through leveraging multiview methodology and local information. Meanwhile, the results show that L-QMVKL exhibits competitive accuracy compared to its classical counterparts. Our work holds promise for advancing the theoretical and practical understanding of quantum kernel methods.

Close

PrevNext

View figure in article

• Machine learning
• Quantum algorithms
• Hybrid model
• Supervised learning