Machine learning algorithms based on generalized Gibbs ensembles

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.

arXiv commentVersion published in J. Stat. Mech. Improved discussion, references added. 16 pages

References in corpus (23)

• Scikit-learn: Machine Learning in Python
• Nonequilibrium dynamics of closed interacting quantum systems
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Relaxation in a Completely Integrable Many-Body Quantum System: An Ab Initio Study of the Dynamics of the Highly Excited States of Lattice Hard-Core Bosons
• Machine learning phases of matter
• Quantum many-body systems out of equilibrium
• Equilibration, thermalisation, and the emergence of statistical mechanics in closed quantum systems
• Many-body quantum state tomography with neural networks
• Generalized Gibbs ensemble in integrable lattice models
• Complete Generalized Gibbs Ensemble in an interacting Theory
• Unsupervised learning of phase transitions: from principal component analysis to variational autoencoders
• Dephasing and the steady state in quantum many-particle systems
• Why does deep and cheap learning work so well?
• Quantum Boltzmann Machine
• Reduced Density Matrix after a Quantum Quench
• Mutual Information, Neural Networks and the Renormalization Group
• Tomography and Generative Data Modeling via Quantum Boltzmann Training
• Detection of phase transition via convolutional neural network
• An exact mapping between the Variational Renormalization Group and Deep Learning
• Machine learning out-of-equilibrium phases of matter
• Zamolodchikov-Faddeev Algebra and Quantum Quenches in Integrable Field Theories
• Deep Learning the Ising Model Near Criticality
• Deep Learning and Quantum Entanglement: Fundamental Connections with Implications to Network Design

Cited by in corpus (2)

• Drawing Phase Diagrams of Random Quantum Systems by Deep Learning the Wave Functions
• An Enquiry on similarities between Renormalization Group and Auto-Encoders using Transfer Learning