Quantum Down Sampling Filter for Variational Auto-encoder

TL;DR

This paper presents a quantum-enhanced variational autoencoder (Q-VAE) that integrates quantum encoding to improve image reconstruction quality, outperforming classical VAEs on standard datasets.

Contribution

Introduction of a hybrid quantum-classical VAE model that leverages quantum encoding within the encoder for better generative performance.

Findings

Q-VAE outperforms classical VAEs on MNIST and USPS datasets.

Q-VAE achieves lower Fréchet inception distance scores.

Q-VAE demonstrates superior image fidelity and reconstruction quality.

Abstract

Variational autoencoders (VAEs) are fundamental for generative modeling and image reconstruction, yet their performance often struggles to maintain high fidelity in reconstructions. This study introduces a hybrid model, quantum variational autoencoder (Q-VAE), which integrates quantum encoding within the encoder while utilizing fully connected layers to extract meaningful representations. The decoder uses transposed convolution layers for up-sampling. The Q-VAE is evaluated against the classical VAE and the classical direct-passing VAE, which utilizes windowed pooling filters. Results on the MNIST and USPS datasets demonstrate that Q-VAE consistently outperforms classical approaches, achieving lower Fr\'echet inception distance scores, thereby indicating superior image fidelity and enhanced reconstruction quality. These findings highlight the potential of Q-VAE for high-quality synthetic data generation and improved image reconstruction in generative models.