Presentation 1998/1/22
Learning of the Way of Abstraction in Cognitive Agents
Atsushi Ueno, Hideaki Takeda, Toyoaki Nishida,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes a method for a cognitive agent to learn the way of abstraction and the policy of behavior selection simultaneously. The main problem of agents in the real environment is how to abstract useful information from a large amount of information in the environment. This is called "the frame problem". We consider that learning of the way of abstraction is a key function for solving the frame problem practically. Our developed system extracts "situations" and maintains them dynamically in the continuous state space on the basis of rewards from the environment. This situation can be regarded as empirically obtained symbol. In this way, the system learns the way of abstraction in a dynamic environment. The results of computer simulations are given.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) symbolization / articulation / frame problem / reinforcement learning / situation classification
Paper # AI97-61
Date of Issue

Conference Information
Committee AI
Conference Date 1998/1/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Learning of the Way of Abstraction in Cognitive Agents
Sub Title (in English)
Keyword(1) symbolization
Keyword(2) articulation
Keyword(3) frame problem
Keyword(4) reinforcement learning
Keyword(5) situation classification
1st Author's Name Atsushi Ueno
1st Author's Affiliation Graduate School of Information Science, Nara Institute of Science and Technology()
2nd Author's Name Hideaki Takeda
2nd Author's Affiliation Graduate School of Information Science, Nara Institute of Science and Technology
3rd Author's Name Toyoaki Nishida
3rd Author's Affiliation Graduate School of Information Science, Nara Institute of Science and Technology
Date 1998/1/22
Paper # AI97-61
Volume (vol) vol.97
Number (no) 498
Page pp.pp.-
#Pages 8
Date of Issue