Quantum Annealing for Variational Bayes Inference
This is an incremental improvement for researchers in machine learning and Bayesian inference, focusing on enhancing optimization in variational methods.
The paper tackles the problem of variational Bayes inference by proposing a quantum annealing-based algorithm (QAVB) that extends simulated annealing for variational Bayes (SAVB), and it shows that QAVB finds better local optima than SAVB in terms of variational free energy in latent Dirichlet allocation (LDA).
This paper presents studies on a deterministic annealing algorithm based on quantum annealing for variational Bayes (QAVB) inference, which can be seen as an extension of the simulated annealing for variational Bayes (SAVB) inference. QAVB is as easy as SAVB to implement. Experiments revealed QAVB finds a better local optimum than SAVB in terms of the variational free energy in latent Dirichlet allocation (LDA).