Distributed Partially Observable Markov Decision Problems (Distributed POMDPs)

TEAMCORE Research Group

Maintained by: Robert Zinkov

Distributed POMDPs are used to model the problem of deriving joint policies that maximize a reward function for a group of agents in a partially observable environment.
If you wish to share your code and the data sets for Distributed POMDP type formalisms on this page, please contact me (zinkov at usc dot edu).

The second workshop on "Multiagent Sequential Decision Making in Uncertain Domains 2007 (MSDM-07)" was held at the 6th international joint conference on Autonomous Agents and Multi Agent Systems (AAMAS-07). More information can be obtained at http://pollux.usc.edu/~maheswar/msdm2007/.

Formalism

References

Code and Datasets

Network Distributed POMDP (ND-POMDP)

         P. Varakantham, J. Marecki, M. Tambe, M. Yokoo. SPIDER attack on a network of POMDPs: Towards quality bounded solutions . In AAAI Spring Symposium,2007.

 

Network Distributed POMDP (ND-POMDP)

         P. Varakantham, J. Marecki, Y. Yabu, M. Tambe, M. Yokoo. Letting loose a SPIDER on a network of POMDPs: Generating quality guaranteed policies . In AAMAS,2007.

 

Network Distributed POMDP (ND-POMDP)

         Y. Kim, R. Nair, P. Varakantham, M. Tambe, M. Yokoo. Exploiting Locality of Interaction in Networked Distributed POMDPs . In AAAI Spring Symposium, 2006.

 

Security through Policy Randomization in Distributed POMDPs

        P. Paruchuri, M. Tambe, F. Ordonez, S. Kraus. Security in multiagent Systems by Policy Randomization . In AAMAS,2006.

 

Continuous Space   Distributed POMDPs (CS-JESP)

         P. Varakantham, R. Nair, M. Tambe, M. Yokoo. Winning back the cup for distributed POMDPs: Planning for continuous belief spaces . In AAMAS,2006.

 

Network Distributed POMDP (ND-POMDP)

         R. Nair, P. Varakantham, M. Tambe, M. Yokoo. Networked Distributed POMDPs: A Synthesis of Distributed Constraint Optimization and POMDPs. In AAAI,2005.

Code, Datasets

Locally Optimal Algorithms for Distributed POMDPs 

        R. Nair, M. Roth, M. Yokoo, M. Tambe. Taming Communication for Improving Policy Computation in Distributed POMDPs. In AAMAS,2004.

        R. Nair, D. Pynadath, M. Yokoo, M. Tambe, S. Marsella. Taming Decentralized POMDPs: Towards Efficient Policy Computation for Multiagent Settings. In IJCAI,2003.

Code

Role-based Multiagent Team Decision Problem

        R. Nair, M. Tambe, S. Marsella. Role allocation and reallocation in multiagent systems: Towards a practical analysis. In AAMAS,2003.

 

Resource Constraints in Team Settings

         P.Paruchuri, M. Tambe, F.Ordonez, S.Kraus. Role allocationTowards a formalization of teamwork with resource constraints. In AAMAS,2003.

 

Communicative MultiAgent Team Decision Problem (COM-MTDP)

         Pynadath, D. and Tambe, M. Multiagent teamwork: Analyzing key teamwork theories and models. First Autonomous Agents and Multiagent Systems Conference (AAMAS), 2002

         Pynadath, D. and Tambe, M. The communicative multiagent team decision problem: Analyzing teamwork theories and models. In Journal of AI Research (JAIR), 2002

Code and Datasets

Relevant Presentations

1.     Presentation on ND-POMDPs at AAAI Spring Symposium, 2007, pdf

2.     Presentation on ND-POMDPs at AAAI, 2005, ppt.

3.     Presentation on Distributed POMDPs at Microsoft Research, 2004, ppt

4.     Presentation on COM-MTDPs at AAMAS, 2002, ppt.