“Self-Organizing Neural Nets and Real Cell-Cell Communication: DEVS Protocol Models at the Brain Circuit Level”
Self-organizing neural nets (SONN) implement certain operations at a macro level that require complex protocols at the individual cell (micro) level. A critical operation assumes that given an input (e.g. facial image), the first and second neurons closest to the input (in recognition space) can identify themselves. This operation is easy to perform for a global algorithm that has access to all the neurons and their locations in space. But we will show that it requires a fairly complex communication protocol for individual cells to follow. We claim that DEVS is the most suitable modeling abstraction to represent such a protocol because individual elements must have states, signals, and memory to properly coordinate in space and time. This and other cases, such as neuron encoding of elapsed time, indicate that current biologically plausible neuron cell models are too simple and incapable of such sophisticated coordination. In this talk, we suggest how DEVS modeling and simulation might resolve questions surrounding brain circuit level realization of cognitive computations.
Bernard P. Zeigler is Professor Emeritus of Electrical and Computer Engineering at the University of Arizona. He received a Ph.D. in Computer/Communication Sciences from the University of Michigan (1968), the M.S. in Electrical Engineering from the Massachusetts Institute of Technology (1964), and a B.S. in Engineering Physics from McGill University in Montreal, Canada (1962). Zeigler has held faculty appointments at the University of Michigan, the Weizmann Institute in Israel, Wayne State University, The University of Arizona, and Arizona State University. While in Arizona, Dr. Zeigler served as the Co-Director of the Arizona Center for Integrative Modeling and Simulation (ACIMS). He is currently affiliated with the Center of Excellence in Command, Control, Communications, Computing and Intelligence (C4I Center) at George Mason University and is also the Chief Scientist at RTSync Corp. Zeigler is best known for his theoretical work concerning modeling and simulation performed according to general systems theory. Zeigler has played a critical role within the modeling and simulation field through his editorial responsibilities and professional activities that focus on the importance of modeling and simulation in its interaction with other disciplines. Zeigler has received much recognition for his various scholarly publications, achievements, and professional service.