Agent architecture based on an interactive self-reflection classifier system
- 56 Downloads
This paper extends the learning classifier system (LCS) to introduce a mechanism for recognizing a current situation by determining a boundary between self and others, and investigates its capability through interaction with an agent. Intensive simulations for adapting to an interacting agent have revealed the following implications: (1) the proposed architecture adapts to an interacting agent more quickly and appropriately than the conventional LCS, and (2) our architecture keeps its adaptation to an interacting agent even when the agent changes its internal models before our architecture acquires it completely.
Key wordsInteractive Self-reflection Boundary setting Learning classifier system Reasoning of user characteristics
Unable to display preview. Download preview PDF.
- 1.Takadama K, Shimohara K (2001) Interactive self-reflection and its architecture based on cellular automata. Artif Life Robotics 5:86–91Google Scholar
- 2.Shiose T, Sawaragi T, Katai O, et al. (2000) Learning is equal to participation: interactive acquisition of identity in a group. 5th International Symposium on Artificial Life and Robotics (AROB '2000), p 434–437Google Scholar
- 3.Yamaguchi T (2000) Interactive self-reflection-based reinforcement learning for multiagent coordination. 5th International Symposium on Artificial Life and Robotics (AROB '2000), p 442–445Google Scholar
- 5.Wilson, SW (1994) ZCS: a zeroth level classifier system. Evol Comput 2:1–18Google Scholar
- 6.Russell SJ, Norving P (1995) Artificial intelligence: a modern approach. Prentice-Hall, Englewood CliffsGoogle Scholar
- 7.Holland JH (1985) Properties of the bucket brigade algorithm. 1st International Conference on Genetic Algorithms (ICGA '85), p 1–7Google Scholar