Training Quantum Boltzmann Machines with Coresets

TL;DR

This paper introduces a coreset-based approach to accelerate training of Quantum Boltzmann Machines by reducing data size, leading to potential practical savings on quantum hardware.

Contribution

It applies coreset techniques to Quantum Boltzmann Machines, reducing training time by minimizing the number of gradient steps needed.

Findings

Coreset approach reduces training time on quantum hardware.

Evaluation on binary image dataset shows improved efficiency.

Potential for significant practical savings in quantum machine learning.

Abstract

Recent work has proposed and explored using coreset techniques for quantum algorithms that operate on classical data sets to accelerate the applicability of these algorithms on near-term quantum devices. We apply these ideas to Quantum Boltzmann Machines (QBM) where gradient-based steps which require Gibbs state sampling are the main computational bottleneck during training. By using a coreset in place of the full data set, we try to minimize the number of steps needed and accelerate the overall training time. In a regime where computational time on quantum computers is a precious resource, we propose this might lead to substantial practical savings. We evaluate this approach on 6x6 binary images from an augmented bars and stripes data set using a QBM with 36 visible units and 8 hidden units. Using an Inception score inspired metric, we compare QBM training times with and without using coresets.