logo
ResearchBunny Logo
Towards provably efficient quantum algorithms for large-scale machine-learning models

Computer Science

Towards provably efficient quantum algorithms for large-scale machine-learning models

J. Liu, M. Liu, et al.

This cutting-edge research by Junyu Liu, Minzhao Liu, Jin-Peng Liu, Ziyu Ye, Yunfei Wang, Yuri Alexeev, Jens Eisert, and Liang Jiang delves into the transformative potential of fault-tolerant quantum computing for training large machine learning models. The authors reveal a quantum algorithm that significantly reduces time complexity, demonstrating promising numerical experiments showcasing quantum enhancements in the training process.

00:00
00:00
Playback language: English
Abstract
This paper explores the potential of fault-tolerant quantum computing to improve the efficiency of training large machine learning models. The authors demonstrate that for sufficiently dissipative and sparse models, a quantum algorithm can achieve a time complexity of O(T² x polylog(n)), where n is the model size and T is the number of training iterations. This is achieved by adapting quantum algorithms for dissipative differential equations to solve the stochastic gradient descent problem. Numerical experiments on models with parameters ranging from 7 million to 103 million show that quantum enhancement is possible in the early stages of training after model pruning, suggesting a sparse parameter download and re-upload scheme.
Publisher
Nature Communications
Published On
Jan 10, 2024
Authors
Junyu Liu, Minzhao Liu, Jin-Peng Liu, Ziyu Ye, Yunfei Wang, Yuri Alexeev, Jens Eisert, Liang Jiang
Tags
fault-tolerant quantum computing
machine learning models
quantum algorithm
stochastic gradient descent
time complexity
numerical experiments
model pruning

Related Publications

Explore these studies to deepen your understanding of the subject.

The Goldilocks paradigm: comparing classical machine learning, large language models, and few-shot learning for drug discovery applications
Computer Science

The Goldilocks paradigm: comparing classical machine learning, large language models, and few-shot learning for drug discovery applications

S. H. Snyder, P. A. Vignaux, et al.

Bayesian Linear Regression for Accurate and Efficient Atomistic Machine Learning Models
Engineering and Technology

Bayesian Linear Regression for Accurate and Efficient Atomistic Machine Learning Models

C. V. D. Oord

Large language models streamline automated machine learning for clinical studies
Medicine and Health

Large language models streamline automated machine learning for clinical studies

S. T. Arasteh, T. Han, et al.

Exploiting redundancy in large materials datasets for efficient machine learning with less data
Engineering and Technology

Exploiting redundancy in large materials datasets for efficient machine learning with less data

K. Li, D. Persaud, et al.

Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny