Andrew Guillory
ABSTRACT
We present a method for learning a human understandable, executable model of an agent's behavior using observations of its interaction with the environment. By executable we mean that the model is suitable for direct execution by an agent. Traditional models of behavior used for recognition tasks (e.g., Hidden Markov Models) are insufficent for this problem because they cannot respond to input from the environment. We train an Input/Output Hidden Markov Model where the output distributions are mixtures of learned low level actions and the transition distributions are conditional on features detected by the agent's sensors. We show that the system is able to learn both the behavior and human-understandable structure of a simulated model.
- R. C. Arkin and T. R. Balch. Aura: Principles and practice in review. Journal of Experimental and Theoretical Artificial Intelligence (JETAI), 9(2--3):175--188, April 1997.Google Scholar
- Y. Bengio and P. Frasconi. Input-output HMM's for sequence processing. IEEE Transactions on Neural Networks, 7(5):1231--1249, September 1996. Google ScholarDigital Library
- D. C. Bentivegna, A. Ude, C. G. Atkeson, and G. Cheng. Humanoid robot learning and game playing using pc-based vision. In 2002 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), volume 3, pages 2449--2454, 2002.Google ScholarCross Ref
- F. Delmotte and M. Egerstedt. Reconstruction of low-complexity control programs from data. In 43rd IEEE Conference on Decision and Control, volume 2, pages 1460--1465, December 2004.Google ScholarCross Ref
- D. Goldberg and M. J. Matarić. Detecting regime changes with a mobile robot using multiple models. In 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS01), volume 2, pages 619--624, 2001.Google ScholarCross Ref
- L. R. Rabiner. A tutorial on hidden markov models and selected applications in speech recognition. Proceedings of the IEEE, 77:257--286, February 1989.Google ScholarCross Ref
- C. Sammu, S. Hurst, D. Kedzier, and D. Michie. Learning to fly. In Ninth International Conference on Machine Learning, 1992. Google ScholarDigital Library
Index Terms
- Learning executable agent behaviors from observation
Recommendations
Learning Sharing Behaviors with Arbitrary Numbers of Agents
AAMAS '18: Proceedings of the 17th International Conference on Autonomous Agents and MultiAgent SystemsWe propose a method for modeling and learning turn-taking behaviors for accessing a shared resource. We model the individual behavior for each agent in an interaction and then use a multi-agent fusion model to generate a summary over the expected ...
An Overview of Opponent Modeling for Multi-agent Competition
Machine Learning for Cyber SecurityAbstractMulti-agent system (MAS) is an area of distributed artificial intelligence that emphasizes the joint behaviors of agents with some degree of autonomy and the complexities arising from their interactions. Multi-agent systems can solve problems that ...
Simulating human behaviors in agent societies
AAMAS '08: Proceedings of the 7th international joint conference on Autonomous agents and multiagent systems - Volume 3As increasing numbers of processors and agents pervade the human environment, societies comprising both humans and agents will emerge. Presently, it is unknown how a person might fair in such mixed societies. For the societies to operate effectively and ...
Comments