Noise-injected analog Ising machines enable ultrafast statistical sampling and machine learning

Ising machines are a promising non-von-Neumann computational concept for neural network training and combinatorial optimization. However, while various neural networks can be implemented with Ising machines, their inability to perform fast statistical sampling makes them inefficient for training neu...

Full description

Saved in:
Bibliographic Details
Published in:Nature communications 2022-10, Vol.13 (1), p.5847-5847, Article 5847
Main Authors: Böhm, Fabian, Alonso-Urquijo, Diego, Verschaffelt, Guy, Van der Sande, Guy
Format: Article
Language:English
Subjects:
639/624/1075/401
639/705/117
639/766/259
639/766/530/2801
Combinatorial analysis
Computer applications
Computers
Digital computers
Humanities and Social Sciences
Ising model
Learning algorithms
Machine learning
multidisciplinary
Neural networks
Optimization
Optoelectronics
Sampling
Science
Science (multidisciplinary)
Software
Statistical sampling
Statistics
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Ising machines are a promising non-von-Neumann computational concept for neural network training and combinatorial optimization. However, while various neural networks can be implemented with Ising machines, their inability to perform fast statistical sampling makes them inefficient for training neural networks compared to digital computers. Here, we introduce a universal concept to achieve ultrafast statistical sampling with analog Ising machines by injecting noise. With an opto-electronic Ising machine, we experimentally demonstrate that this can be used for accurate sampling of Boltzmann distributions and for unsupervised training of neural networks, with equal accuracy as software-based training. Through simulations, we find that Ising machines can perform statistical sampling orders-of-magnitudes faster than software-based methods. This enables the use of Ising machines beyond combinatorial optimization and makes them into efficient tools for machine learning and other applications. Ising machines are accelerators for computing difficult optimization problems. In this work, Böhm et al. demonstrate a method that extends their use to perform statistical sampling and machine learning orders-of-magnitudes faster than digital computers.
ISSN:2041-1723
2041-1723
DOI:10.1038/s41467-022-33441-3