Algorithmic knowledge profiles for introspective monitoring in artificial cognitive agents
| Autor: | Adan A. Gomez, Juan C. Giraldo, Manuel F. Caro |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Cognitive systems
Computer science business.industry media_common.quotation_subject Representation (systemics) Cognition 02 engineering and technology Hafnium compounds Planner Visualization 03 medical and health sciences 0302 clinical medicine 0202 electrical engineering electronic engineering information engineering Introspection 020201 artificial intelligence & image processing Artificial intelligence business computer 030217 neurology & neurosurgery computer.programming_language media_common |
| Zdroj: | IEEE ICCI*CC |
| DOI: | 10.1109/icci-cc.2017.8109792 |
| Popis: | This paper describes a new approach for the representation of profiles of cognitive functions. The profiles are used in introspective monitoring to keep updated the meta-level about the most relevant attributes of each cognitive function that is executed in the object-level. The profiles are called algorithmic knowledge profiles. Declarative meta-knowledge, procedural meta-knowledge and strategic meta-knowledge are represented using the algorithmic knowledge profile. The profiles are formally defined and the computational implementation is based on extended tabular trees. The approach is illustrated using an instructional planner agent as an example. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|