Cooperative cellular transformation systems


Sanjay S. Deshpande, Lev Goldfarb and Virendra C. Bhavsar


Order and its emergence in natural systems is a profound phenomenon and is being studied by almost all sciences. We attempt to outline our initial understanding of some aspects of the topic as it relates to "intelligent" systems. We sketch a new cooperative computational architecture for realizing the inductive learning process, based on the recently proposed evolving transformation systems (ETS) model. The outlined architecture is based on cooperative cellular organization of computation and draws on analogy with the biological intelligent systems. More accurately, we attempt to capture the phenomenon of emergence of order in the process of concept formation during inductive learning in a two fold manner: (abstract) inductive class order via ETS and architectural class order via CCTS (Cooperative Cellular Transformation System). The proposed architectural model of intelligent processes, CCTS, provides a fundamentally new insight into the understanding of neurobiological aspects of information processing. It suggests a close relationship between the chemical form of computation seen in biological systems and the symbolic cognitive functions that they realize. Thus, the main objective of this paper is to attract researchers interested in the development of intelligent architectures to experiment with various truly cooperative architectures, the need for which becomes so acute with the realization of the advantages offered by the Evolving Transformation System model as compared with the very popular artificial neural network model.
last updated: 95/12/22