Natural policy gradient methods and the covariance matrix adaptation evolution strategy, two variable metric methods proposed for solving reinforcement learning tasks, are contrast...
As computational learning agents move into domains that incur real costs (e.g., autonomous driving or financial investment), it will be necessary to learn good policies without n...
In this paper, we describe methods for e ciently computing better solutions to control problems in continuous state spaces. We provide algorithms that exploit online search to boo...
Bayesian priors offer a compact yet general means of incorporating domain knowledge into many learning tasks. The correctness of the Bayesian analysis and inference, however, lar...
In real world scenes, objects to be classified are usually not visible from every direction, since they are almost always positioned on some kind of opaque plane. When moving a cam...