Machine learning algorithms based on generalized Gibbs ensembles

DOI10.1088/1742-5468/AAE025MaRDI QIDQ3303207zbMATH OpenOpenAlexWikidataFDO

Authors Tatjana Puškarov, Axel Cortés Cubero

Publication date 11 August 2020

Published in Journal of Statistical Mechanics: Theory and Experiment (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1804.03546

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Quantum equilibrium statistical mechanics (general) (82B10)

Abstract: Machine learning algorithms often take inspiration from established results and knowledge from statistical physics. A prototypical example is the Boltzmann machine algorithm for supervised learning, which utilizes knowledge of classical thermal partition functions and the Boltzmann distribution. Recently, a quantum version of the Boltzmann machine was introduced by Amin, et. al., however, non-commutativity of quantum operators renders the training process by minimizing a cost function inefficient. Recent advances in the study of non-equilibrium quantum integrable systems, which never thermalize, have lead to the exploration of a wider class of statistical ensembles. These systems may be described by the so-called generalized Gibbs ensemble (GGE), which incorporates a number of "effective temperatures". We propose that these GGE's can be successfully applied as the basis of a Boltzmann-machine-like learning algorithm, which operates by learning the optimal values of effective temperatures. We show that the GGE algorithm is an optimal quantum Boltzmann machine: it is the only quantum machine that circumvents the quantum training-process problem. We apply a simplified version of the GGE algorithm, where quantum effects are suppressed, to the classification of handwritten digits in the MNIST database. While lower error rates can be found with other state-of-the-art algorithms, we find that our algorithm reaches relatively low error rates while learning a much smaller number of parameters than would be needed in a traditional Boltzmann machine, thereby reducing computational cost.

Recommendations

• Restricted Boltzmann machine based on a Fermi sea
• Learning quantum models from quantum or classical data
• Quantum machine learning
• Quantum learning algorithms for quantum measurements
• Quantum machine learning: a classical perspective

Cites work

• scientific article; zbMATH DE number 6276119 (Why is no real title available?)
• A Fast Learning Algorithm for Deep Belief Nets
• Deep learning
• Generalized Gibbs ensemble in integrable lattice models
• Neural networks and physical systems with emergent collective computational abilities
• Reducing the Dimensionality of Data with Neural Networks
• Scikit-learn: machine learning in Python
• Solving the quantum many-body problem with artificial neural networks
• Statistical field theory. An introduction to exactly solved models in statistical physics.
• The generalized Gibbs ensemble for Heisenberg spin chains
• Training Products of Experts by Minimizing Contrastive Divergence
• Two soluble models of an antiferromagnetic chain
• Why does deep and cheap learning work so well?
• Zamolodchikov-Faddeev algebra and quantum quenches in integrable field theories

Cited in

(1)

• Restricted Boltzmann machine based on a Fermi sea

Describes a project that uses

Uses Software

• STL-10 dataset
• darch
• Scikit

This page was built for publication: Machine learning algorithms based on generalized Gibbs ensembles

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3303207)