Doran Chakraborty's Homepage
Publications
My Open Source Software
TAC page
Learning Agent Research Group (LARG)
My wife's blog
Personal Information
I am a PhD candidate at the Computer Science Department of the University Of Texas, at Austin. I am advised by Peter Stone .
My research involves modeling and learning agent behavior in multiagent environments. Large scale deployments of multiagent systems calls for the need of better learners that come with better guarantees, than the guarantees provided by the current state of the art learners. Through my research, I hope to contribute to the problem at a level which will benefit the field, and more so, the world. To this end, my main research agenda is to develop Multiagent learning algorithms, that will come with the best possible guarantees (justifiable from both theoretical and empirical stand points) against arbitrary opponents, in game theoretic settings.
I also have a parallel interest in structure learning in factored reinforcement learning problems, where the problem is to identify the structure of the underlying factored domain and then to subsequently converge to near optimal behavior in tractable number of steps.
News
November 2010: I just completed my PhD proposal. The title of my dissertation is " Learning in the Presence of Other Agents". I would be more than happy to exchange ideas on some of the problems that I have identified in my dissertation. In this regard feel free to drop in an email, if any problem interests you, or appears related to your area of research.
July 2009: David Pardoe, me, and Peter Stone, won the first Trading Agent Competition on Ad Auctions with our agent TacTex-2009. The competition was held at the 21'st International Joint Conference on Artificial Intelligence, Pasadena, USA.
Contact Information
mailto: chakrado at cs dot utexas dot edu, doran dot chakraborty at gmail dot com