KaBaGe-RL: Kanerva-based generalisation and reinforcement learning for possession football

IEEE International Conference on Intelligent Robots and Systems

Volumn 1, Issue , 2001, Pages 292-297

  Kostiadis, Kostas ^a   Hu, Huosheng ^a  

^a UNIVERSITY OF ESSEX   (United Kingdom)

Author keywords

Decision making; Generalisation; Kanerva coding; Reinforcement learning; RoboCup

Indexed keywords

APPROXIMATION THEORY; DECISION MAKING; KNOWLEDGE REPRESENTATION; NEURAL NETWORKS; OPTIMIZATION;

FEATURE VECTOR; KANERVA CODING; REINFORCEMENT LEARNING; ROBOCUP;

ROBOT LEARNING;

EID: 0035558808     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (23)

1. Residual algorithms: Reinforcement learning with function approximation
2. Artificial neural networks and approximate reasoning for intelligent control in space
3. Multivariable functional interpolation and adaptive networks
4. Distributed representations
5. Essex Wizards 2000 Team Description
6. Reinforcement learning: A survey
7. Robocup: The robot world cup initiative
8. A multi-threaded approach to simulated soccer agents for the robocup competition
9. The use of design patterns for the development of multi-agent systems
10. A self-learning rule-based controller employing approximate reasoning and neutral net concepts
11. Soccer server: A tool for research on multiagent systems
12. Karlsruhe brainstormers - A reinforcement learning approach to robotic soccer
13. Reinforcement learning for 3 vs. 2 keepaway
14. Generalisation in reinforcement learning: Successful examples using sparse coarse coding
15. Online learning with random representations
16. Vision-based reinforcement learning for robocup: Towards real robot competition
17. A heuristic approach to reinforcement learning control systems