Scaling Up Optimal Heuristic Search in Dec-POMDPs via Incremental Expansion

By Matthijs T. J. Spaan, Frans Adriaan Oliehoek, Chris Amato. 2011

55 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

How researchers see this paper

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

In Artificial Intelligence

1. Shimon Whiteson
2. Chris Amato
3. Frans Adriaan Oliehoek
4. Matthijs T. J. Spaan
5. Stefan J. Witwicki
6. Francisco S. Melo
7. Bikramjit Banerjee
8. Landon Kraemer

In Computational Theory and Mathematics

1. Shimon Whiteson
2. Frans Adriaan Oliehoek
3. Matthijs T. J. Spaan
4. Francisco S. Melo
5. Stefan J. Witwicki
6. Bikramjit Banerjee
7. Landon Kraemer
8. Chris Amato

Citations in the corpus

• Multi-agent reinforcement learning as a rehearsal for decentralized planning (2016, critical): “Many exact solvers have been developed for Dec-POMDPs [28,19,26], but these”
• Monte-Carlo Expectation Maximization for Decentralized POMDPs (2013, unknown): “To explore the policy space, the simplest technique is using the -greedy strategy [Sutton and Barto, 1998].”
• Concurrent reinforcement learning as a rehearsal for decentralized planning under uncertainty (2013, unknown): “[3], but these approaches are not scalable because the underlying problem is provably NEXP-complete [2].”
• Sample Bounded Distributed Reinforcement Learning for Decentralized POMDPs (2012, methodology): “Because solving Dec-POMDPs is NEXP-complete (Bernstein et al. 2002), exact solu-tion techniques for finite horizon problems require signifi-cant time and memory resources (Szer and Charpillet 2006; Oliehoek et al. 2010; Spaan, Oliehoek, and Amato 2011).”
• A Concise Introduction to Decentralized POMDPs (2016, background): “, 2009], or to try and avoid the expansion of all child nodes by incrementally expanding nodes only when needed [Spaan et al., 2011].”
• Exploiting Agent and Type Independence in Collaborative Graphical Bayesian Games (2011, methodology): “the source problems. For problems small enough to solve optimally, we compare the solution quality of FACTORED FSPC to that of GMAA*-ICE, the state-of-the-art method for optimally solving Dec-POMDPs (Spaan et al., 2011). We also compare against several other approximate methods for solving Dec-POMDP”
• QueryPOMDP: POMDP-Based Communication in Multiagent Systems (2011, methodology): “For comparison, we also provide the optimal value for the 6-step horizon, reported in [25].”