Title :
Interaction, evolution, and intelligence
Author :
Goldin, Dina ; Keil, David
Author_Institution :
Dept. of Math. & Comput. Sci., Massachusetts Univ., Boston, MA, USA
Abstract :
Evolutionary computation (EC) and intelligence are closely related. Adaptability can be considered an indispensable aspect of intelligence; in turn, interaction of an agent with its environment is necessary for adaptation. Interaction is fundamental to EC at multiple levels, and an interactive viewpoint enhances our understanding of EC principles. Interaction is based on a stream-based rather than string-based view of computation, modeled mathematically by coinduction and coalgebras. It forces us to rethink some fundamental assumptions of computer science. The challenge of building intelligent systems can be better met by accepting a paradigm shift from traditional algorithmic models of computation toward interaction, with all its implications
Keywords :
evolutionary computation; coalgebras; coinduction; evolution; evolutionary computation; intelligence; intelligent systems; interaction; Computational modeling; Computer networks; Computer science; Concurrent computing; Embedded computing; Intelligent agent; Mathematical model; Mathematics; Military computing; Physics computing;
Conference_Titel :
Evolutionary Computation, 2001. Proceedings of the 2001 Congress on
Conference_Location :
Seoul
Print_ISBN :
0-7803-6657-3
DOI :
10.1109/CEC.2001.934273