Quantum-Train: Rethinking Hybrid Quantum-Classical Machine Learning in the Model Compression Perspective
Preprint
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.
Keywords
All Related Versions
This publication has 46 references indexed in Scilit:
- The complexity of NISQNature Communications, 2023
- Theoretical error performance analysis for variational quantum circuit based functional regressionnpj Quantum Information, 2023
- Path sampling of recurrent neural networks by incorporating known physicsNature Communications, 2022
- Generalization in quantum machine learning from few training dataNature Communications, 2022
- A Grover-search based quantum learning scheme for classificationNew Journal of Physics, 2021
- Learning molecular dynamics with simple language model built upon long short-term memory neural networkNature Communications, 2020
- Machine learning the thermodynamic arrow of timeNature Physics, 2020
- Quantum Chemistry in the Age of Quantum ComputingChemical Reviews, 2019
- Learning phase transitions by confusionNature Physics, 2017
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989