A Framework for Behavioural Cloning

This paper describes recent experiments in automatically constructing reactive agents. The method used is behavioural cloning, where the logged data from skilled, human operators are input to an induction program which outputs a control strategy for a complex control task. Initial studies were able to successfully construct such behavioural clones, but suered from several drawbacks, namely, that the clones were brittle and dicult to understand. Current research is aimed at solving these problems by learning in a framework where there is a separation between an agent’s goals and its knowledge of how to achieve them.

[1]  A. Whitehead.  An Introduction to Mathematics , 1949, Nature.

[2]  Claude Semmut.  Concept learning by experiment , 1981, IJCAI 1981.

[3]  Claude Sammut.  Concept Learning by Experiment , 1981, IJCAI.

[4]  Donald Michie,et al.  The superarticulacy phenomenon in the context of software manufacture , 1990 .

[5]  Marcel Schoppers,et al.  Universal Plans for Reactive Robots in Unpredictable Environments , 1987, IJCAI.

[6]  Paul Compton,et al.  Knowledge in Context: A Strategy for Expert System Maintenance , 1990, Australian Joint Conference on Artificial Intelligence.

[7]  Paolo Mancarella,et al.  Generalized Stable Models: A Semantics for Abduction , 1990, ECAI.

[8]  Donald Michie,et al.  Cognitive models from subcognitive skills , 1990 .

[9]  Céline Rouveirol,et al.  Beyond Inversion of Resolution , 1990, ML.

[10]  Claude Sammut,et al.  Controlling a Black-Box Simulation of a Spacecraft , 1991, AI Mag..

[11]  Claude Sammut,et al.  Learning to Fly , 1992, ML.

[12]  Rui Camacho,et al.  Building symbolic representations of intuitive real-time skills from performance data , 1994, Machine Intelligence 13.

[13]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[14]  Nils J. Nilsson,et al.  Reacting, Planning, and Learning in an Autonomous Agent , 1996, Machine Intelligence 14.

[15]  Nils J. Nilsson,et al.  Teleo-Reactive Programs for Agent Control , 1993, J. Artif. Intell. Res..

[16]  Ivan Bratko,et al.  Reconstructing Human Skill with Machine Learning , 1994, European Conference on Artificial Intelligence.

[17]  Robert A. Kowalski,et al.  Using Meta-Logic to Reconcile Reactive with Rational Agents , 1995, PAAM.

[18]  De Raedt,et al.  Advances in Inductive Logic Programming , 1996 .

[19]  Scott Sherwood Benson,et al.  Learning action models for reactive autonomous agents , 1996 .

[20]  Claude Sammut,et al.  Combining Knowledge Acquisition and Machine Learning to Control Dynamic Systems , 1997, IJCAI.