Investigation of D-Wave quantum annealing for training Restricted Boltzmann Machines and mitigating catastrophic forgetting

TL;DR

This paper investigates the use of D-Wave quantum annealing for training Restricted Boltzmann Machines and explores its potential in mitigating catastrophic forgetting, finding limited benefits for training but promising applications in generative replay.

Contribution

It introduces a hybrid sampling method combining classical and quantum approaches and demonstrates the first use of QA-generated patterns for catastrophic forgetting mitigation.

Findings

No significant improvement in RBM training with QA sampling.

QA can generate diverse samples from low-probability regions.

Potential of QA in generative replay for continual learning.

Abstract

Modest statistical differences between the sampling performances of the D-Wave quantum annealer (QA) and the classical Markov Chain Monte Carlo (MCMC), when applied to Restricted Boltzmann Machines (RBMs), are explored to explain, and possibly address, the absence of significant and consistent improvements in RBM trainability when the D-Wave sampling was used in previous investigations. A novel hybrid sampling approach, combining the classical and the QA contributions, is investigated as a promising way to benefit from the modest differences between the two sampling methods. No improvements in the RBM training are achieved in this work, thereby suggesting that the differences between the QA-based and MCMC sampling, mainly found in the medium-to-low probability regions of the distribution, which are less important for the quality of the sample, are insufficient to benefit the training. Difficulties in achieving sufficiently high quality of embedding RBMs into the lattice of the newer generation of D-Wave hardware could be further complicating the task. On the other hand, the ability to generate samples of sufficient variety from lower-probability parts of the distribution has a potential to benefit other machine learning applications, such as the mitigation of catastrophic forgetting (CF) during incremental learning. The feasibility of using QA-generated patterns of desirable classes for CF mitigation by the generative replay is demonstrated in this work for the first time. While the efficiency of the CF mitigation using the D-Wave QA was comparable to that of the classical mitigation, both the speed of generating a large number of distinct desirable patterns and the potential for further improvement make this approach promising for a variety of challenging machine learning applications.