On the Challenges of Physical Implementations of RBMs

Abstract

Restricted Boltzmann machines (RBMs) are powerful machine learning models, but learning and some kinds of inference in the model require sampling-based approximations, which, in classical digital computers, are implemented using expensive MCMC. Physical computation offers the opportunity to reduce the cost of sampling by building physical systems whose natural dynamics correspond to drawing samples from the desired RBM distribution. Such a system avoids the burn-in and mixing cost of a Markov chain. However, hardware implementations of this variety usually entail limitations such as low-precision and limited range of the parameters and restrictions on the size and topology of the RBM. We conduct software simulations to determine how harmful each of these restrictions is. Our simulations are designed to reproduce aspects of the D-Wave quantum computer, but the issues we investigate arise in most forms of physical computation.

References in corpus (1)

• Improving neural networks by preventing co-adaptation of feature detectors

Cited by in corpus (6)

• Quantum machine learning: a classical perspective
• Opportunities and challenges for quantum-assisted machine learning in near-term quantum computers
• Application of Quantum Annealing to Training of Deep Neural Networks
• A Boltzmann Machine Implementation for the D-Wave
• Quantum-inspired annealers as Boltzmann generators for machine learning and statistical physics
• Towards Sampling from Nondirected Probabilistic Graphical models using a D-Wave Quantum Annealer