A Case-Based Reasoning Framework for Developing Agents Using Learning by Observation

Most realistic environments are complex, partially observable and impose real-time constraints on agents operating within them. This paper describes a framework that allows agents to learn by observation in such environments. When learning by observation, agents observe an expert performing a task and learn to perform the same task based on those observations. Our framework aims to allow agents to learn in a variety of domains (physical or virtual) regardless of the behaviour or goals of the observed expert. To achieve this we ensure that there is a clear separation between the central reasoning system and any domain-specific information. We present case studies in the domains of obstacle avoidance, robotic arm control, simulated soccer and Tetris.

[1]  Ian D. Watson,et al.  Similarity-Based Retrieval and Solution Re-use Policies in the Game of Texas Hold'em , 2010, ICCBR.

[2]  Barbara Messing,et al.  An Introduction to MultiAgent Systems , 2002, Künstliche Intell..

[3]  Christian Bauckhage,et al.  Combining Self Organizing Maps and Multilayer Perceptrons to Learn Bot-Behavior for a Commercial Computer Game , 2003 .

[4]  A. Ram,et al.  Authoring Behaviors for Games using Learning from Demonstration , 2009 .

[5]  Santiago Ontañón,et al.  Case-Based Reasoning and User-Generated Artificial Intelligence for Real-Time Strategy Games , 2011 .

[6]  Babak Esfandiari,et al.  Supplemental Case Acquisition Using Mixed-Initiative Control , 2011, FLAIRS Conference.

[7]  Pedro A. González-Calero,et al.  Building CBR systems with jcolibri , 2007, Sci. Comput. Program..

[8]  Edwina L. Rissland,et al.  Case Retrieval through Multiple Indexing and Heuristic Search , 1993, IJCAI.

[9]  Padraig Cunningham,et al.  Case-Based Plan Recognition in Computer Games , 2003, ICCBR.

[10]  D.H. Grollman,et al.  Learning robot soccer skills from demonstration , 2007, 2007 IEEE 6th International Conference on Development and Learning.

[11]  Babak Esfandiari,et al.  An Active Approach to Automatic Case Generation , 2009, ICCBR.

[12]  Santiago Ontañón,et al.  MMPM: a Generic Platform for Case-Based Planning Research , 2010, CGAMES 2010.

[13]  Agnar Aamodt,et al.  Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches , 1994, AI Commun..

[14]  Luc Lamontagne,et al.  Forgetting Reinforced Cases , 2008, ECCBR.

[15]  Santiago Ontañón,et al.  Case-Based Planning and Execution for Real-Time Strategy Games , 2007, ICCBR.

[16]  Ralph Johnson,et al.  design patterns elements of reusable object oriented software , 2019 .

[17]  Christian Bauckhage,et al.  Combining Self Organizing Maps and Multilayer Perceptrons to Learn Bot-Behaviour for a Commercial Game , 2003, GAME-ON.

[18]  Jonathan Dinerstein,et al.  DEMONSTRATION‐BASED BEHAVIOR PROGRAMMING FOR EMBODIED VIRTUAL AGENTS , 2008, Comput. Intell..

[19]  T. Roth-Berghofer,et al.  Rapid Prototyping of CBR Applications with the Open Source Tool myCBR , 2008 .

[20]  Mark T. Keane,et al.  On the Automatic Generation of Cases Libraries by Chunking Chess Games , 1995, ICCBR.

[21]  Babak Esfandiari,et al.  Considerations for Real-Time Spatially-Aware Case-Based Reasoning: A Case Study in Robotic Soccer Imitation , 2008, ECCBR.

[22]  Hector Muñoz-Avila,et al.  Imitating Inscrutable Enemies: Learning from Stochastic Policy Observation, Retrieval and Reuse , 2010, ICCBR.

[23]  Babak Esfandiari,et al.  A Case-Based Reasoning Approach to Imitating RoboCup Players , 2008, FLAIRS.

[24]  Stefan Schaal,et al.  Robot Learning From Demonstration , 1997, ICML.

[25]  Pieter Abbeel,et al.  Learning for control from multiple demonstrations , 2008, ICML '08.