We present a method for learning a human understandable, executable model of an agent's behavior using observations of its interaction with the environment. By executable we ...
Andrew Guillory, Hai Nguyen, Tucker R. Balch, Char...
We provide a novel view of learning an approximate model of a partially observable environment from data and present a simple implemenf the idea. The learned model abstracts away ...
We apply the auxiliary particle filter algorithm of Pitt and Shephard (1999) to the problem of robot localization. To deal with the high-dimensional sensor observations (images) ...
We present a novel technique to exploit the power-performance tradeoff. The technique can be used stand-alone or in conjunction with dynamic voltage scaling, the mainstream techn...
This paper examines the problem of finding an optimal policy for a Partially Observable Markov Decision Process (POMDP) when the model is not known or is only poorly specified. W...