Channel Attention for Quantum Convolutional Neural Networks
arXiv:2311.02871 · doi:10.1103/PhysRevA.110.012447
Abstract
Quantum convolutional neural networks (QCNNs) have gathered attention as one of the most promising algorithms for quantum machine learning. Reduction in the cost of training as well as improvement in performance is required for practical implementation of these models. In this study, we propose a channel attention mechanism for QCNNs and show the effectiveness of this approach for quantum phase classification problems. Our attention mechanism creates multiple channels of output state based on measurement of quantum bits. This simple approach improves the performance of QCNNs and outperforms a conventional approach using feedforward neural networks as the additional post-processing.
6 pages, 4 figures, 1 table
References in corpus (12)
- Variational Quantum Algorithms
- Bloch vectors for qudits
- Beyond Barren Plateaus: Quantum Variational Algorithms Are Swamped With Traps
- One-Dimensional Symmetry Protected Topological Phases and their Transitions
- Application-Oriented Performance Benchmarks for Quantum Computing
- Realizing Quantum Convolutional Neural Networks on a Superconducting Quantum Processor to Recognize Quantum Phases
- Barren plateaus in quantum tensor network optimization
- Analyzing the barren plateau phenomenon in training quantum neural networks with the ZX-calculus
- Quantum phase detection generalisation from marginal quantum neural network models
- Model-Independent Learning of Quantum Phases of Matter with Quantum Convolutional Neural Networks
- Testing platform-independent quantum error mitigation on noisy quantum computers
- Randomness-enhanced expressivity of quantum neural networks