Exploiting Anonymity and Homogeneity in Factored Dec-MDPs through Precomputed Binomial Distributions
Speaker (s):
Rajiv Ranjan KUMAR
Research Engineer
School of Information Systems
Singapore Management University
|
|
Date:
Time:
Venue:
|
|
May 5, 2017, Friday
2:30 pm - 3:00 pm
Meeting Room 4.4, Level 4
School of Information Systems
Singapore Management University
80 Stamford Road
Singapore 178902
We look forward to seeing you at this research seminar.
|
|
About the Talk
Recent work in decentralized stochastic planning for cooperative agents has focused on exploiting homogeneity of agents and anonymity in interactions to solve problems with large numbers of agents. Due to a linear optimization formulation that computes joint policy and an objective that indirectly approximates joint expected reward with reward for expected number of agents in all state, action pairs, these approaches have ensured improved scalability. Such an objective closely approximates joint expected reward when there are many agents, due to law of large numbers. However, the performance deteriorates in problems with fewer agents. In this paper, we improve on the previous line of work by providing a linear optimization formulation that employs a more direct approximation of joint expected reward. The new approximation is based on offline computation of binomial distributions. Our new technique is not only able to improve quality performance on problems with large numbers of agents, but is able to perform on par with existing best approaches on problems with fewer agents. This is achieved without sacrificing on scalability/run-time performance of previous work.
This a pre-conference talk for Sixteenth International Conference on Antonomous Agents and Multiagent Sytems (AAMAS -2017).
About the Speaker
Rajiv Ranjan KUMAR is a research engineer in School of Information Systems, Singapore Management University and working under the supervision of Associate Professor Pradeep Varakantham. He received his B.Tech in Computer Science & Engineering from Kalinga Institute of Industrial Technology(KIIT), Bhubaneswar, India and M.Tech in Computer Science Engineering from Visvesvaraya National Institute of Technology(VNIT), Nagpur, India. He then worked as a software engineer in Persistent Systems in Nagpur, India. He works in the area of Intelligent Systems & Decision Analytics (ISDA). His key research interest lies in decision support in uncertain environments.
