Title: Turing on Super-Turing
Speaker: Hava Siegelmann, University of Massachusetts, Amherst
Date: Monday, November 19, 2012 11:00am - 12:00pm
Location: DIMACS Center, CoRE Bldg, Room 431, Rutgers University, Busch Campus, Piscataway, NJ
Abstract:
Alan Turing's research, following his introduction of the Logical Universal Model and predating universal computers, sought a different form of computation that would more closely simulate the brain and achieve intelligence: "My contention is that machines can be constructed which will simulate the behaviour of the human mind very closely."
Over the years, Turing suggested various attributes necessary to achieving this goal: Learning from experience, neural elements that develop connections based on environmental interaction, guidance by reward or external teacher, random elements, reading the bits of pi, etc. He pointed to the brain both as a refutation of the relevance of Godel-like theorems as a limit to machine intelligence and in support of a different, superior form of computation.
Turing's was a singular mind; his writings from 1936-1938 are the basis of the computer revolution, Computer Science and Artificial Intelligence; a fresh look at his work from 1947-1953, in light of his quest for intelligent machines, may provide relevant insights into computer intelligence and shed light on biological processing.
The parallels between Turing's predictions and Super-Turing computational theory are striking. We explain Super-Turing computation, its development in 1993 at Rutgers, its relation to natural computation, and discuss current efforts to harness this model toward a deeper comprehension of intelligence and computation in biological systems.
DIMACS/CCICADA Interdisciplinary Series, Complete Fall Calendar 2012