— While the Partially Observable Markov Decision Process (POMDP) provides a formal framework for the problem of robot control under uncertainty, it typically assumes a known and ...
Markov Decision Processes are a powerful framework for planning under uncertainty, but current algorithms have difficulties scaling to large problems. We present a novel probabil...
The search for finite-state controllers for partially observable Markov decision processes (POMDPs) is often based on approaches like gradient ascent, attractive because of their ...
For a Markov Decision Process with finite state (size S) and action spaces (size A per state), we propose a new algorithm--Delayed Q-Learning. We prove it is PAC, achieving near o...
Alexander L. Strehl, Lihong Li, Eric Wiewiora, Joh...
The algorithmic framework developed for improving heuristic solutions of the new version of deterministic TSP [Choi et al., 2002] is extended to the stochastic case. To verify the...