Expectation-maximization Methods for Solving (po)mdps and Optimal Control Problems

By Marc Toussaint, Amos J. Storkey, Stefan Harmeling. 2010

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Sergey Levine
2. Yoshua Bengio
3. Pieter Abbeel
4. Jan Peters
5. Chelsea Finn
6. Jens Kober
7. J. Urgen Schmidhuber
8. Trevor Darrell

Citations in the corpus

• Reinforcement learning in robotics: A survey (2013, background): “, 1999), (ii) policy updates inspired by expectation– maximization (EM) (Toussaint et al., 2010), and (iii) the path integral methods (Kappen, 2005).”
• On the Computational Complexity of Stochastic Controller Optimization in POMDPs (2011, background): “INTRODUCTION Partially observable Markov decision processes (POMDPs) have proven to be a valu­able conceptual tool for problems throughout AI, including reinforcement learning [Chrisman 1992], planning under uncertainty [Kaelbling et al. 1998], and multiagent coordination [Bernstein et al. 2005].”