PhD Dissertation Proposal by Rajiv Ranjan KUMAR | Towards Improving System Performance in Large Scale Multi-Agent Systems with Selfish Agent

Rapid development in artificial intelligence (AI) technologies (e.g., Multi-Agent System Planning, Algorithmic Game Theory, Deep Reinforcement Learning, etc.) has made them very popular for optimizing and improving large scale multi-agent systems. Such technologies are more desirable when dealing with selfish agents in the system that are not willing to cooperate among themselves. This work is targeted at providing solution approaches for improving large scale multi-agent systems with selfish agents. In such systems, the performance of an agent depends on not just his/her own efforts, but also on other agent's decisions. This complex interactions among multiple agents, coupled with the large scale of the problem domain and the uncertainties from environment, makes decision making very challenging. In this work we study the problem where centralized aggregator is present, that wants to maximize the revenue of entire system. To this end, in this work we study this problem from strategic and operational point of view. More precisely, from strategic decision making, we propose planning and deep reinforcement learning based solution algorithms to improve the system performance by optimizing the adaptive operating hours of selfish agents. From operational point of view, we propose deep learning-based solution algorithm to motivate selfish agents, to induce cooperation among themselves. Basically, through strategic and operational decision making, we assist selfish agents in making intelligent decisions that results in improved system performance.

Rajiv Ranjan Kumar is a Ph.D. candidate in the School of Computing and Information Systems, Singapore Management University, advised by Associate Professor Pradeep Varakantham. He received his B.Tech in Computer Science & Engineering from KIIT Bhubaneswar, India and M.Tech in Computer Science Engineering from NIT, Nagpur, India. He then worked as a software engineer at Persistent Systems Limited, Nagpur, before joining as a research engineer in SMU in 2013. His key research interests include Decision Making under Uncertainty, Deep Reinforcement Learning and Multi-agent Systems. His PhD work is focused on techniques for improving system performance in large scale multi-agent systems with selfish agents.