A meta-learning method based on temporal difference error

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volumn 5863 LNCS, Issue PART 1, 2009, Pages 530-537

  Kobayashi, Kunikazu ^a   Mizoue, Hiroyuki ^a   Kuremoto, Takashi ^a   Obayashi, Masanao ^a  

^a YAMAGUCHI UNIVERSITY   (Japan)

Author keywords

Inverted pendulum control problem; Maze search problem; Meta learning; Meta parameter; Reinforcement learning; TD error

Indexed keywords

DISCOUNT RATES; EXTERNAL ENVIRONMENTS; INVERTED PENDULUM; LEARNING RATES; META-PARAMETER; METALEARNING; NEUROMODULATORS; SEARCH PROBLEM; TEMPORAL DIFFERENCE ERRORS; TEMPORAL DIFFERENCES;

COMPUTER SIMULATION; DATA PROCESSING; LEARNING ALGORITHMS; REINFORCEMENT; REINFORCEMENT LEARNING;

PENDULUMS;

EID: 76649092973     PISSN: 03029743     EISSN: 16113349     Source Type: Book Series    
DOI: 10.1007/978-3-642-10677-4_60     Document Type: Conference Paper

References (6)

1. Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT Press, Cambridge (1998)
2. Schultz, W., Dayan, P., Montague, P.R.: A Neural Substrate of Prediction and Reward. Science 275, 1593-1599 (1997)
3. Doya, K.: Metalearning and Neuromodulation. Neural Networks 15, 495-506 (2002)
4. Schweighofer, N., Doya, K.: Meta-learning in Reinforcement Learning. Neural Networks 16(1), 5-9 (2003)
5. Watkins, C.J.C.H., Dayan, P.: Q-learning. Machine Learning 8(3-4), 279-292 (1992)
6. Ishii, S., Yoshida, W., Yoshimoto, J.: Control of Exploitation- Exploration Meta-parameter in Reinforcement Learning. Neural Networks 15(4-6), 665-687 (2002)