Multi-Class Quantum Convolutional Neural Networks
This work addresses classification problems in information retrieval, but it is incremental as it applies an existing quantum method to a new multi-class scenario.
The authors tackled multi-class classification of classical data by proposing a quantum convolutional neural network (QCNN), which was tested on the MNIST dataset with varying numbers of classes; the results showed that the QCNN slightly underperformed classical CNNs with 4 classes but outperformed them with higher numbers of classes.
Classification is particularly relevant to Information Retrieval, as it is used in various subtasks of the search pipeline. In this work, we propose a quantum convolutional neural network (QCNN) for multi-class classification of classical data. The model is implemented using PennyLane. The optimization process is conducted by minimizing the cross-entropy loss through parameterized quantum circuit optimization. The QCNN is tested on the MNIST dataset with 4, 6, 8 and 10 classes. The results show that with 4 classes, the performance is slightly lower compared to the classical CNN, while with a higher number of classes, the QCNN outperforms the classical neural network.