Batch-iFDD for representation expansion in large MDPs
Matching pursuit (MP) methods are a promising class of feature construction algorithms for value function approximation. Yet existing MP methods require creating a pool of potential features, mandating expert knowledge or enumeration of a large feature pool, both of which hinder scalability. This pa...
Main Authors: | Geramifard, Alborz, Walsh, Thomas J., Roy, Nicholas, How, Jonathan P. |
---|---|
Other Authors: | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
Format: | Article |
Language: | en_US |
Published: |
Association for Uncertainty in Artificial Intelligence (AUAI)
2015
|
Online Access: | http://hdl.handle.net/1721.1/97035 https://orcid.org/0000-0001-8576-1930 https://orcid.org/0000-0002-2508-1957 https://orcid.org/0000-0002-8293-0492 |
Similar Items
-
Practical reinforcement learning using representation learning and safe exploration for large scale Markov decision processes
by: Geramifard, Alborz, 1980-
Published: (2012) -
Combining dynamic abstractions in large MDPs
by: Steinkraus, Kurt, et al.
Published: (2005) -
UAV Cooperative Control with Stochastic Risk Models
by: Geramifard, Alborz, et al.
Published: (2013) -
Markov decision processes in artificial intelligence : MDPs, beyond MDPs and applications /
by: Sigaud, Olivier, et al.
Published: (2010) -
Transience in countable MDPs
by: Kiefer, SM, et al.
Published: (2021)