Learning Better Trading Dialogue Policies by Inferring Opponent Preferences

Ioannis Efstathiou, Oliver Lemon

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Negotiation dialogue capabilities have been identified as important in a variety of application areas. In prior work, it was shown how Reinforcement Learning (RL) agents can learn to use implicit and explicit manipulation moves in dialogue to manipulate their adversaries in non-cooperative trading games. We now show that trading dialogues are more successful when the RL agent builds an opponent model -- an estimate of the (hidden) goals and preferences of the adversary -- and learns how to exploit them. We explore a variety of state space representations for the preferences of trading adversaries, including one based on Conditional Preference Networks (CP-NETS), used for the first time in RL. We show that representing adversary preferences leads to significant improvements in trading success rates.
Original languageEnglish
Title of host publicationProceedings of AAMAS 2016
Pages1403-1404
Number of pages2
Publication statusPublished - 9 May 2016
Event15th International Conference on Autonomous Agents and Multiagent Systems 2016 - Singapore, Singapore, Singapore
Duration: 9 May 201613 May 2016

Conference

Conference15th International Conference on Autonomous Agents and Multiagent Systems 2016
Abbreviated titleAAMAS 2016
CountrySingapore
CitySingapore
Period9/05/1613/05/16

Fingerprint Dive into the research topics of 'Learning Better Trading Dialogue Policies by Inferring Opponent Preferences'. Together they form a unique fingerprint.

  • Cite this