Ekhlas Sonu
University of Georgia
12 Papers
46 Citations
Ekhlas Sonu is an academic researcher from University of Georgia. The author has contributed to research in topics: Partially observable Markov decision process & Markov decision process. The author has an hindex of 6, co-authored 12 publications. Previous affiliations of Ekhlas Sonu include Stanford University.
Chat about Author
Papers
Scalable solutions of interactive POMDPs using generalized and bounded policy iteration
Ekhlas Sonu,Prashant Doshi +1 more
TL;DR: This article shows how the bounded policy iteration technique may be performed with anytime behavior in settings formalized by the interactive POMDP framework, and extensively evaluates the approach on multiple problem domains with some that are significantly large in their dimensions and in contexts with uncertainty about the other agent’s frames and those involving multiple other agents.
27
Decision-Theoretic Planning Under Anonymity in Agent Populations
TL;DR: A method for drastically scaling the finitely-nested I-POMDP to certain agent populations for the first time, and a comprehensive empirical evaluation of the methods on three new problems domains -- policing large protests, controlling traffic congestion at a busy intersection, and improving the AI for the popular Clash of Clans multiplayer game.
•Posted Content
Individual Planning in Agent Populations: Exploiting Anonymity and Frame-Action Hypergraphs
TL;DR: In this paper, the authors model and extend anonymity and context-specific independence for computational gain in multi-agent decision-making in a partially observable Markov decision process (I-POMDP) setting.
12
Generalized and bounded policy iteration for finitely-nested interactive POMDPs: scaling up
Ekhlas Sonu,Prashant Doshi +1 more
- 04 Jun 2012
TL;DR: This paper shows how the bounded policy iteration technique may be performed in settings formalized by the interactive POMDP framework, and evaluates the approach on multiple problem domains, and demonstrates its properties and scalability.
GaTAC: a scalable and realistic testbed for multiagent decision making (demonstration)
Ekhlas Sonu,Prashant Doshi +1 more
- 04 Jun 2012
TL;DR: GaTAC provides a low-cost, open-source and flexible environment for realistically simulating and evaluating policies generated by multi-agent decision making algorithms in real world problem domains pertaining to control of autonomous uninhabbited aerial vehicles (AUAVs).