| Summary: | This thesis introduces a novel way of adapting the Hybrid Monte Carlo (HMC) algorithm using Gaussian process bandits. HMC is a powerful Markov chain Monte Carlo (MCMC) method, but it requires careful tuning of its hyper-parameters. We propose a Gaussian process bandit approach to carry out the adaptation of the hyper-parameters while the Markov chain progresses. We also introduce the use of cross-validation error measures for adaptation, which we believe are more pragmatic than many existing adaptation objectives. The new measures take the intended statistical use of the model, whose parameters are estimated by HMC, into consideration. We apply these two innovations to the adaptation of HMC for prediction and feature selection with multi-layer feed-forward neural networks. The experiments with synthetic and real data show that the proposed adaptive scheme is not only automatic, but also does better tuning than human experts.
|
|---|
