Q(st at):= (I Ã¢â¬â o')Q(st at) + o'(r(st+1)
AbstractStraightforward reinforcement learning for multi-agent co-learning settings often results in poor outcomes. Meta-learning processes beyond straightforward reinforcement learning may be necessary to achieve good (or optimal) outcomes. Algorithmic processes of meta-learning, or "manipulation", will be described, which is a cognitively realistic and effective means for learning cooperation. We will discuss various "manipulation" routines that address the issue of improving multi-agent co-learning. We hope to develop better adaptive means of multi-agent cooperation, without requiring a priori knowledge, and advance multi-agent co-learning beyond existing theories and techniques
Similar books and articles
An Evolutionary Game Theoretic Perspective on Learning in Multi-Agent Systems.Karl Tuyls, Ann Nowe, Tom Lenaerts & Bernard Manderick - 2004 - Synthese 139 (2):297 - 330.
Individual Action and Collective Function: From Sociology to Multi-Agent Learning.Ron Sun - manuscript
Spontaneous Coordination and Evolutionary Learning Processes in an Agent-Based Model.Pierre Barbaroux & Gilles Enée - 2005 - Mind and Society 4 (2):179-195.
A Model for Updates in a Multi-Agent Setting.John Cantwell - 2007 - Journal of Applied Non-Classical Logics 17 (2):183-196.
Beyond Simple Rule Extraction: The Extraction of Planning Knowledge From Reinforcement Learners.Ron Sun - unknown
Bottom-Up Skill Learning in Reactive Sequential Decision Tasks.Ron Sun, Todd Peterson & Edward Merrill - unknown
Learning to Cooperate: Reciprocity and Self-Control.Peter Danielson - 2002 - Behavioral and Brain Sciences 25 (2):256-257.
Learning with Neighbours: Emergence of Convention in a Society of Learning Agents.Roland Mühlenbernd - 2011 - Synthese 183 (S1):87-109.
Added to PP
Historical graph of downloads