Training Quantum Boltzmann Machines with Coresets
This work addresses training efficiency for quantum machine learning models, but it is incremental as it applies existing coreset techniques to a specific quantum algorithm.
The authors tackled the computational bottleneck of Gibbs state sampling in Quantum Boltzmann Machines by using coresets to reduce training steps, achieving a reduction in training time on a 6x6 binary image dataset.
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.