A Functional Model of Neuronal Networks
Dr. James Lo
Department of Mathematics and Statistics
University of
Maryland, Baltimore County
| Time: noon-1 pm | Location: MP 401 |
Abstract:
In his 1958 book, The Computer and the Brain, John von Neumann
said: "We require exquisite numerical precision over many logical steps to
achieve what brains accomplish in very few short steps." After more than 50
years, the natural question, "What are those very few short steps?" still
remains to be fully answered.
This talk presents a functional model of neuronal networks resulting from my
attempt to answer the question. The model, called probabilistic associative
memory (PAM), has many functions that have not been explained satisfactorily
or jointly by other models. The purpose of the talk is to solicit comments
and suggestions on PAM and stimulate discussions on the above question.
PAM is ready for testing as a learning machine for application to
recognition and understanding of handwriting, face, speech, radiograph,
license plate, video, etc.