On the challenges of using D-Wave computers to sample Boltzmann Random Variables
This work highlights incremental difficulties in applying quantum computing to a specific machine learning problem, with implications for researchers in quantum algorithms and neural networks.
The paper addresses the challenges of using D-Wave quantum computers to sample Boltzmann random variables, an NP-hard problem relevant for applications like training Boltzmann machines, and finds that significant obstacles remain for efficient sampling.
Sampling random variables following a Boltzmann distribution is an NP-hard problem involved in various applications such as training of \textit{Boltzmann machines}, a specific kind of neural network. Several attempts have been made to use a D-Wave quantum computer to sample such a distribution, as this could lead to significant speedup in these applications. Yet, at present, several challenges remain to efficiently perform such sampling. We detail the various obstacles and explain the remaining difficulties in solving the sampling problem on a D-wave machine.