🤖 AI Summary
This study investigates the feasibility of quantum computing for MNIST handwritten digit classification. To address the challenge of encoding high-dimensional classical image data into shallow quantum circuits, we propose an end-to-end quantum-classical hybrid framework: a classical autoencoder first compresses input images to 64-dimensional latent representations; these are then encoded into a 5-qubit parameterized quantum circuit—comprising RX, RZ, H, and CNOT gates—for quantum feature mapping and measurement. Measurement outcomes undergo differentiable sampling before being fed into a fully connected classifier with BatchNorm and Dropout. We introduce, for the first time on MNIST, a transfer learning paradigm grounded in quantum measurement outputs and establish a fully differentiable quantum-classical co-training pipeline. The model achieves 94.2% test accuracy—slightly below the classical baseline (96.7%)—yet demonstrates that quantum intermediate representations exhibit semantic separability and compatibility with downstream classical models, providing both a novel paradigm and empirical validation for quantum-enhanced machine learning on NISQ devices.
📝 Abstract
In this research, we explore the integration of quantum computing with classical machine learning for image classification tasks, specifically focusing on the MNIST dataset. We propose a hybrid quantum-classical approach that leverages the strengths of both paradigms. The process begins with preprocessing the MNIST dataset, normalizing the pixel values, and reshaping the images into vectors. An autoencoder compresses these 784-dimensional vectors into a 64-dimensional latent space, effectively reducing the data's dimensionality while preserving essential features. These compressed features are then processed using a quantum circuit implemented on a 5-qubit system. The quantum circuit applies rotation gates based on the feature values, followed by Hadamard and CNOT gates to entangle the qubits, and measurements are taken to generate quantum outcomes. These outcomes serve as input for a classical neural network designed to classify the MNIST digits. The classical neural network comprises multiple dense layers with batch normalization and dropout to enhance generalization and performance. We evaluate the performance of this hybrid model and compare it with a purely classical approach. The experimental results indicate that while the hybrid model demonstrates the feasibility of integrating quantum computing with classical techniques, the accuracy of the final model, trained on quantum outcomes, is currently lower than the classical model trained on compressed features. This research highlights the potential of quantum computing in machine learning, though further optimization and advanced quantum algorithms are necessary to achieve superior performance.