Cognitive modeling with context sensitive reinforcement learning
Author
Editor
- J. Malek
Summary, in English
We describe how a standard reinforcement learning algorithm can be changed to include a second contextual input that is used to modulate the learning in the original algorithm. The new algorithm takes the context into account during relearning when the previously learned actions are no longer valid. The algorithm was tested on a number of cognitive experiment and shown to reproduce the learning in both a task switching test and in the Wisconsin Card Sorting Test. In addition, the algorithm was able to learn a context sensitive categorization of objects in the Labov experiment.
Department/s
Publishing year
2004
Language
English
Pages
10-19
Publication/Series
Proceedings of AILS 04 ( Report / Lund Institute of Technology, Lund University ; 151)
Full text
- Available as PDF - 183 kB
- Download statistics
Links
Document type
Conference paper
Publisher
Department of Computer Science, Lund University
Topic
- Computer Vision and Robotics (Autonomous Systems)
Status
Published
Project
- Ikaros: An infrastructure for system level modelling of the brain
ISBN/ISSN/Other
- ISSN: 1650-1276