Display Publications by [Year] [Type] [Topic]
Multiagent teams have been shown to be effective in many domains that require coordination among team members. However, find- ing valuable joint-actions becomes increasingly difficult in tightly- coupled domains where each agentŐs performance depends on the actions of many other agents. Reward shaping partially addresses this challenge by deriving more Ňtuned" rewards to provide agents with additional feedback, but this approach still relies on agents randomly discovering suitable joint-actions. In this work, we introduce Counterfactual Agent Suggestions (CAS) as a method for injecting knowledge into an agentŐs learning process within the confines of existing reward structures. We show that CAS enables agent teams to converge towards desired behaviors more reliably. We also show that improvement in team performance in the presence of suggestions extends to large teams and tightly-coupled domains.
@InProceedings{tumer-zerbel_aamas20,
author = {and K. Tumer},
title = {The Power of Suggestion},
booktitle = {Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems},
address = {Auckland, New Zealand},
month = {May},
pages={},
abstract={Multiagent teams have been shown to be effective in many domains that require coordination among team members. However, find- ing valuable joint-actions becomes increasingly difficult in tightly- coupled domains where each agentÕs performance depends on the actions of many other agents. Reward shaping partially addresses this challenge by deriving more Òtuned" rewards to provide agents with additional feedback, but this approach still relies on agents randomly discovering suitable joint-actions. In this work, we introduce Counterfactual Agent Suggestions (CAS) as a method for injecting knowledge into an agentÕs learning process within the confines of existing reward structures. We show that CAS enables agent teams to converge towards desired behaviors more reliably. We also show that improvement in team performance in the presence of suggestions extends to large teams and tightly-coupled domains.},
bib2html_pubtype = {Refereed Conference Papers},
bib2html_rescat = {Multiagent Systems, Evolutionary Algorithms},
year = {2020}
}
Generated by bib2html.pl (written by Patrick Riley ) on Wed Apr 01, 2020 17:39:43