Bengisu Guresti

I am currently a Computer Science PhD student at Washington University in St. Louis and my research is on mechanism design for multi-agent learning. I am interested in deep reinforcement learning, multi-agent systems and game theory.

Email  /  CV  /  Google Scholar  /  GitHub /  LinkedIn

I am interested in deep reinforcement learning, multi-agent systems and game theory.

In this work, we propose Incentive Q-Flow (IQ-Flow) algorithm to design incentive mechanisms for increasing social welfare and promoting cooperation. IQ-Flow aims to make the cooperative policy correspond to the self-interested policy of the agents by changing system’s reward setup.

We analyze the evaluation performance for each combination of agent count for training versus evaluation. We perform experimental evaluations on predator prey and traffic junction environments and demonstrate that it is possible to obtain similar or higher evaluation performance by training with less agents.

Robotics is a promising interdisciplinary field with application areas that free humans of manual labor or high risk services. In order to build robots that have the capability of taking jobs that require manual labor or services of high risk and operating in the world, it is essential to provide the robots with the necessary manipulation skills.

This ubiquitous CS researcher website template spawned from here.
Last updated: December 2021