Leveraging Adiabatic Quantum Computation for Election Forecasting
This work addresses election forecasting, a domain-specific application, but appears incremental as it builds on existing quantum methods for Boltzmann machines without demonstrating new breakthroughs.
The paper tackles the problem of sampling from large fully-connected graphs with arbitrary correlations, which becomes intractable as graph size increases, by exploring adiabatic quantum computation for training Boltzmann machines to predict the 2016 Presidential election, but no concrete results or numbers are provided.
Accurate, reliable sampling from fully-connected graphs with arbitrary correlations is a difficult problem. Such sampling requires knowledge of the probabilities of observing every possible state of a graph. As graph size grows, the number of model states becomes intractably large and efficient computation requires full sampling be replaced with heuristics and algorithms that are only approximations of full sampling. This work investigates the potential impact of adiabatic quantum computation for sampling purposes, building on recent successes training Boltzmann machines using a quantum device. We investigate the use case of quantum computation to train Boltzmann machines for predicting the 2016 Presidential election.