PEGASUS: A policy search method for large MDPs and POMDPs
UAI '00 Proceedings of the 16th Conference on Uncertainty in Artificial Intelligence
Monte Carlo Strategies in Scientific Computing
Monte Carlo Strategies in Scientific Computing
Hi-index | 0.00 |
Many inference and optimization tasks in machine learning can be solved by sampling approaches such as Markov Chain Monte Carlo (MCMC) and simulated annealing. These methods can be slow if a single target density query requires many runs of a simulation (or a complete sweep of a training data set). We introduce a hierarchy of MCMC samplers that allow most steps to be taken in the solution space using only a small sample of simulation runs (or training examples). This is shown to accelerate learning in a policy search optimization task.