An approach to the design of reinforcement functions in real world, agent-based applications

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics

Volumn 31, Issue 3, 2001, Pages 288-301

  Bonarini, Andrea ^a,b   Bonacina, Claudio ^b   Matteucci, Matteo ^b  

^a IEEE   (Italy)

^b POLITECNICO DI MILANO   (Italy)

Author keywords

Autonomous robots; Intelligent robots; Learning classifier systems; Mobile robots; Reinforcement learning

Indexed keywords

REINFORCEMENT LEARNING (RL);

COMPUTER SIMULATION; INTELLIGENT ROBOTS; LEARNING ALGORITHMS; MOBILE ROBOTS;

LEARNING SYSTEMS;

EID: 0035359275     PISSN: 10834419     EISSN: None     Source Type: Journal    
DOI: 10.1109/3477.931510     Document Type: Article

References (30)

1. Reinforcement learning: A survey
2. Q-learning
3. TD(λ) converges with probability 1
4. Classifier systems and genetic algorithms
5. Classifier fitness based on accuracy
6. Spatial planning: A configuration space approach
7. Rewarding behaviors
8. The role of the trainer in reinforcement learning
9. Reward functions for accelerated learning
10. Reinforcement learning in the multi-robot domain
11. Learning to predict by the method of temporal differences
12. Reinforcement learning with replacing eligibility traces
13. Anytime learning and adaptation of hierarchical fuzzy logic behaviors
14. Purposive behavior acquisition on a real robot by a vision based reinforcement learning
15. Learning to perceive and act by trial and error
16. Fuzzy and crisp representation of real-valued input for learning classifier systems
17. Efficient exploration in reinforcement learning with hidden state
18. Fuzzy and crisp representation of real-valued input for learning classifier systems
19. Comparing reinforcement learning algorithms applied to crisp and fuzzy learning classifier systems
20. Reinforcement distribution to fuzzy classifiers: A methodology to extend crisp algorithms
21. Evolutionary learning of fuzzy rules: Competition and cooperation
22. Dynamic update of the reinforcement function during learning
23. Learning fuzzy classifier systems