About Me
I am a Postdoctoral Fellow at the Vector Institute, where I am supervised by Professor Pascal Poupart and Professor Sheila McIlraith. My primary research interest is in the area of multi-agent systems. I am interested in the issues of scale, non-stationarity, effective communication, and sample inefficiency in multi-agent learning systems. As a consequence, my area of work is at the intersection of Reinforcement Learning and Game Theory. I am also highly interested in the theoretical aspects of Reinforcement Learning. I have two important and related long-term research goals. The first is to bridge the widening gap between the theoretical understanding and empirical advances of Multi-agent Reinforcement Learning. The second is to make multi-agent learning algorithms applicable to a variety of large-scale real-world problems.
Before the postdoctoral appointment, I was a doctoral student in the Department of Electrical and Computer Engineering (ECE) at the University of Waterloo, advised by Professors Mark Crowley and Kate Larson. As part of my doctoral research, I actively collaborated with Professor Pascal Poupart, Professor Matthew E. Taylor, Professor Issac Tamblyn, and Professor Colin Bellinger. Previously, I obtained a Master’s degree in the ECE department at the University of Waterloo, supervised by Professor Mark Crowley. My bachelor’s degree was in the field of Geomatics Engineering, from College of Engineering, Guindy (CEG), Anna University, Chennai, India (CEG is Asia’s oldest technical institution, and the oldest technical institution outside Europe). During this time, I completed a Mitacs funded research internship in the Spatial Lab, under the supervision of Professor Colin Robertson in Wilfrid Laurier University. I have also worked as a research intern at Borealis AI – Edmonton and Waterloo labs. In general, my research is motivated by Computational Sustainability. During my Ph.D., I was a part of the Vector Postgraduate Affiliate program.
I am highly passionate about increasing equity and diversity across all levels in Canadian institutions of higher learning. I am currently a mentor at the IBET Phd Project to help with this.
News
Sep 2022: I joined the Vector Institute as a Postdoctoral Fellow.
January 2023: Two papers (one full paper and one extended abstract) accepted to AAMAS-2023. See you in London in June if you are attending AAMAS.
April 2023: Co-teaching CS486/CS 686: Introduction to AI at the University of Waterloo. Refer to the course website here.
May 2023: My PhD Dissertation won the CAIAC best doctoral dissertation award. Here is the Twitter announcement. I am giving a talk in Canadian AI in Montreal on my doctoral dissertation on June 8th.
May 2023: Presenting our JAIR paper on Multi-agent advising in IJCAI-2023 (journal track) in Macao, China. See you in Macao in August if you are attending IJCAI.
August 2023: One paper accepted in TMLR. This paper explores the advantages of policy-gradient based learning by leveraging instructions/help from multiple teachers in the same environment.
October 2023: ChemGymRL is accepted to be presented in NeurIPS-23 AI for Accelerated Materials Design Workshop and AI for Science Workshop. See you in New Orleans in December if you are attending NeurIPS.
February 2024: Vector Institute has moved to the Schwartz Reisman Innovation Campus in downtown Toronto.
February 2024: ChemGymRL is accepted to be published in Digital Discovery.
May 2024: One paper accepted to ICML. This paper provides a way of estimating confidence while learning constraints from expert datasets in the context of reinforcement learning. See you in Vienna, Austria if you are attending ICML in July. Full version of the paper can be found in arXiv.