Quantum-Train: Rethinking Hybrid Quantum-Classical Machine Learning in the Model Compression Perspective

TL;DR

Quantum-Train (QT) introduces a hybrid quantum-classical framework that significantly reduces model parameters and enhances efficiency in machine learning tasks, despite slight accuracy trade-offs, leveraging quantum neural networks and classical mappings.

Contribution

QT presents a novel hybrid quantum-classical approach that reduces model size and improves efficiency in machine learning through quantum neural networks and classical mappings.

Findings

Reduces parameter count from M to O(polylog M)

Maintains competitive classification accuracy

Enhances model efficiency and generalization

Abstract

We introduces the Quantum-Train(QT) framework, a novel approach that integrates quantum computing with classical machine learning algorithms to address significant challenges in data encoding, model compression, and inference hardware requirements. Even with a slight decrease in accuracy, QT achieves remarkable results by employing a quantum neural network alongside a classical mapping model, which significantly reduces the parameter count from $M$ to $O(\text{polylog} (M))$ during training. Our experiments demonstrate QT's effectiveness in classification tasks, offering insights into its potential to revolutionize machine learning by leveraging quantum computational advantages. This approach not only improves model efficiency but also reduces generalization errors, showcasing QT's potential across various machine learning applications.