Dr. Prashant Doshi
Director, THINC Lab
THINC lab is located in the Boyd Graduate Studies Research Center in the southern part of the UGA campus. Currently, more than 10 graduate students including doctoral and masters students are conducting research in the lab. Research conducted in the lab has been funded by grants from the Air Force and Army Research Offices, National Science Foundation, National Institutes of Health and the industry.
We present a new framework, OntoNLQA, for querying RDF data annotated using ontologies that allows posing questions in natural language. It introduces a novel approach for discovering the sophisticated semantic associations that may exist between the important entities of a question, in order to build an intuitive query and retrieve precise answers. We apply this framework to the context of parasite immunology data, leading to a system called AskCuebee that allows parasitologists to pose genomic, proteomic and pathway questions in natural language related to the parasite, Trypanosoma cruzi.[Learn more..]
 Kenneth Bogert, Sina Solaimanpour and Prashant Doshi, Aerial Robotic Simulations for Evaluation of Multi-Agent Planning in GaTAC (demonstration), in AAMAS, 2015. [video]
 Xia Qu and Prashant Doshi, On the Role of Fairness and Limited Backward Induction in Sequential Bargaining Games: New Behavioral Models and Analyses, accepted with minor revisions in Annals of Mathematics and AI, Springer, 2015.
 Yingke Chen, Prashant Doshi and Yifeng Zeng, Iterative Online Planning in Multiagent Settings with Limited Model Spaces and PAC Guarantees, full paper in AAMAS, 2015. [paper]
 Kenneth Bogert and Prashant Doshi, Multi-Robot Inverse Reinforcement Learning with State Transition Estimation, extended abstract in AAMAS, 2015.
 Xia Qu and Prashant Doshi, Improved Planning for Infinite-Horizon Interactive POMDPs using Probabilistic Inference, extended abstract in AAMAS, 2015.
 Ekhlas Sonu, Yingke Chen and Prashant Doshi, Individual Planning in Agent Populations: Exploiting Anonymity and Frame-Action Hypergraphs, accepted in ICAPS, 2015.
 Fadel Adoe, Yingke Chen and Prashant Doshi, Fast Solving of Influence Diagrams for Multiagent Planning on GPU-enabled Architectures, full paper in ICAART, 2015. [paper]
 C. Kim, D. Kim, J. Yuan, E. Melias, P. Doshi and R. Hill, Towards Example-Based Learning and Engagement of Teachers in RoboSTEM, in Annual Meeting of American Educational Research Association, 2015.
[03/07/15] We are looking to hire a postdoctoral associate to begin May 15, 2015. Interested applicants should read this.
[1/18/15] Prof. Doshi is on faculty leave of absence at the University of Waterloo beginning Spring 2015 until November 30, 2015.
[7/28/14] Doshi is on the Expert's Panel in the first MIPC workshop at AAAI 2014.
[7/19/14] Doshi and Qiu visited Prof. Makoto Yokoo at Kyushu University and Prof. Michihiro Kandori at Tokyo University in a week long collaborative research visit.
[6/19/14] Prof. Doshi gave a hosted talk on using influence diagrams for decision making in multiagent settings at the CUNY graduate center and visited Google NYC.
[6/6/14] Prof. Doshi is a Co-PI on a grant awarded through UGA's inaugural Interdisicplinary Proposal Development Grant Program. Only six teams out of about 50 were funded. Prof. Doshi's project headed by Prof. ChanMin Kim in the College of Education will focu on RoboSTEM: a collection of open educational resources to help elementary school teachers teach STEM subjects through robotics and design-based learning.
[5/15/14] Yingke Chen PhD AAU joins the THINC lab as a postdoctoral research associate and will supervise projects on multiagent systems.