Title :
Learning in reactive sequential decision tasks: the CLARION model
Author :
Sun, Ron ; Peterson, Todd
Author_Institution :
Dept. of Comput. Sci., Alabama Univ., Tuscaloosa, AL, USA
Abstract :
In order to develop versatile agents that learn in situated contexts and generalize resulting knowledge to different environments, we explore the possibility of learning both procedural and declarative knowledge in a hybrid connectionist architecture. The architecture, CLARION, is based on the two-level idea proposed earlier by the authors. The architecture integrates reactive routines, rules, learning, and decision-making in a unified framework, and structures different learning components synergistically
Keywords :
adaptive systems; generalisation (artificial intelligence); knowledge based systems; learning (artificial intelligence); neural net architecture; CLARION model; adaptive rule induction; agents; decision-making; declarative knowledge; generalization; hybrid connectionist architecture; procedural knowledge; reactive routines; reinforcement learning; Computer architecture; Computer science; Context modeling; Decision making; Encoding; Navigation; Sun; Supervised learning;
Conference_Titel :
Neural Networks, 1996., IEEE International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-3210-5
DOI :
10.1109/ICNN.1996.549047
