Potential-based shaping and Q-value initialization are equivalent

Journal of Artificial Intelligence Research

Volumn 19, Issue , 2003, Pages 205-208

  Wiewiora, Eric ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

COMPUTATIONAL METHODS; FUNCTIONS; LEARNING ALGORITHMS;

Q-VALUES; REINFORCEMENT LEARNING; SHAPING ALGORITHM;

LEARNING SYSTEMS;

EID: 27344453198     PISSN: 10769757     EISSN: 10769757     Source Type: Journal    
DOI: 10.1613/jair.1190     Document Type: Article

References (4)

1. Bertsekas, D. P., & Tsitsiklis, J. T. (1996). Neuro-dynamic Programming. Athena Scientific.
2. Koenig, S., & Simmons, R. (1996). The effect of representation and knowledge on goal-directed exploration with reinforcement-learning algorithms. Machine Learning, 22(1/3), 227-250.
3. Ng, A. Y., Harada, D., & Russell, S. (1999). Policy invariance under reward transformations: theory and application to reward shaping. In Machine Learning, Proceedings of the Sixteenth International Conference, pp. 278-287. Morgan Kaufmann.
4. Sutton, R. S., & Barto, A. G. (1998). Reinforcement Learning: An Introduction. The MIT Press.