Transcript slides - Seidenberg School of Computer Science and Information

Alan M. Turing
“Alan M. Turing and Intelligent Machinery, a
prelude to Artificial Intelligence”
May 4, 2012
Version 2.0; 05/04/2012
John M. Casarella
Proceedings of Student/Faculty Research Day
Ivan G. Seidenberg School of CSIS, Pace University
Introduction
Alan M. Turing:
Mathematician
Computer
Scientist
Cryptographer
The Enigma
Human Being
Turing - On Computable Numbers…




"On computable numbers, with an application to the
Entscheidungsproblem" was published in 1936
Introduces the world to his “computing machine”, complete with
paper tape, symbols, scanner (reader) and of course
“instructions” (program to manage changes in state)
Sample Program Table: State
Scanned Square
Operation
Next State
a
blank
P[0], R
b
b
blank
R
c
c
blank
P[1], R
d
d
blank
R
a
Top Contributions to Modern Computing:
 The Idea of controlling the function of the computing
machinery by storing a program of symbolically encoded
instructions in the machines memory.
 By demonstration, a single machine of fixed structure
capable of completing all computations - the Universal Turing
Machine.
Turing - Computing Machinery
“Computing Machinery and Intelligence” (Mind)
was published in 1950
 Within the paper, the famous question is asked:
Can Machines Think?
 Provides the basis for providing the answer
 Christened the “Turing Test” at a point in the
future
 Much has been afforded the test and much has
been placed on its significance

The Turing Test






Critics ask if passing the test is sufficient or a
necessary condition for machine intelligence
Although widely accepted, limiting in determining if a
machine is capable of intelligence
Turing never claimed passing the is a necessary
condition for intelligence
In his papers, claims point of test was determine if a
computer can “imitate a brain”
Can it be passed at all?
If “machine intelligence” no longer a oxymoron, then
one of Turing’s predictions has come true
In the beginning…


Turing was harboring thoughts of Machine
Intelligence as early as 1941
Turing’s computing machines - model a child’s mind
and then ‘educate’ it






Learning from experience
Start with an initial state of the mind / computer
Determine the education subjected to
Experiences other than education
Start with a simple machine, progress to one more
elaborate
A key concept - “teach” a network of artificial
neurons to perform specific tasks, i.e., the machine
learns
Turing - Intelligent Machinery




“Intelligent Machinery” (Unpublished report, 1948)
Remained unpublished until 1968 (Well after Hebb and
Rosenblatt)
Introduces many of the concepts that were later to
become central to AI, specifically to Neural Nets
The envisioned picture of the cortex as an unorganized
machine is satisfactory from the point of view of
evolution and genetics
Turing and AI Foundations







Computing machines built out of simple, neuron-like
elements
Elements randomly connected together into networks
Consisted of artificial neurons and devices capable of
modifying the connections between them
Training process renders certain pathways as effective
or ineffective
Every neuron executes the same logical operation of
“not and’ (NAND)
The idea that an initially unorganized neural network
can be organized by means of “interference training” this is significant
Referred to as “unorganized machines” or “B-type
unorganized machine” neural net
Turing’s Unorganized Machine
Neurons
Unorganized
Machines
Turing’s Unorganized Machine
Perceptron Model
Unorganized Machines
B - Node
The Perception of Intelligence



How do we perceive intelligence?
What if a problem is presented to a
mathematician or scientist to solve…
What if a problem is presented to a
computer to solve…
Turing’s Perceptions


Turing (1946): there are indications however that it is possible
to make the machine display intelligence at the risk of its
making occasional serious mistakes
Turing (1947): but the human mathematician would likewise
make blunders when trying out new techniques…in other
words, if a machine is expected to be infallible, it cannot also
be intelligent.
Intelligent humans, even highly regarded, intelligent humans are
not infallible; they make errors, yet we do not consider them
any less intelligent when they do, so why not apply the same
standard or perception to computing machining when we
attempt to determine machine intelligence.
And in the end…


His personal expression to be “human” led to
condemnation
On 9/11/2009 the British Publicly Apologize and
acknowledge Turing’s contribution to the War effort
and for providing the foundation for modern
computing
References
[1] A. M. Turing, "Computing Machinery and Intelligence," Mind, vol. 59, pp. 433 - 460, October 1950.
[2] B. J. Copeland, and D. Proudfoot, "What Turing Did after He Invented the Universal Turing Machine," Journal of Logic, Language,
and Information, vol. 9, pp. 491-509, 2000.
[3] B. J. Copeland, and D. Proudfoot, "The Legacy of Alan Turing," Mind, vol. 108, pp. 187-195, 1999.
[4] B. J. Copeland, "The Essential Turing," Oxford, Great Britain: Oxford University Press, 2004.
[5] A. M. Turing, "The Turing Digital Archive," http://www.turingarchive.org/: University of Southamption and King's College
Cambridge, 2002.
[6] N. Block, "Psychologism and Behaviourism," Philosophical Review, vol. 90, pp. 5-43, 1981.
[7] R. French, "Subcognition and the Limits of the Turing Test," Mind, vol. 99, 1990.
[8] P. Hayes, and K. Ford, "Turing Test Considered Harmful," in Proceedings of the Fourteenth International Joint Conference on
Artificial Intelligence, 1995, pp. 972-977.
[9] K. M. Ford, and P.J. Hayes, "On Computational Wings: Rethinking the Goals of Artificial Intelligence," Scientific American Presents,
vol. 9, pp. 78-83, 1998.
[10] J. Copeland and Diane Proudfoot "On Alan Turing's Anticipation of Connectionism," Synthese, vol. 108, pp. 361 - 377, 1996.
[11] J. Copeland and Diane Proudfoot, "Alan Turing's Forgotten Ideas in Computer Science," Scientific American, pp. 99 - 103, 1999.
[12] A. M. Turing, "Intelligent Machinery," in Machine Intelligence 5, B. Meltzer, and D. Michie, Ed. Edinburgh: Edinburgh University
Press, 1948, pp. 3-23.
[13]B. G. Farley, and W.A. Clark, "Simulation of Self-Organizing Systems by Digital Computer," Institute of Radio Enigneers
Transactions on Information Theory, vol. 4, pp. 76 - 84, 1954.
[14]D. Gelernter, "Artificial Intelligence is Lost in the Woods," in Technology Review, 2007.
[15]A. M. Turing, "On computable numbers, with an application to the Entscheidungsproblem," Proceedings of the London Mathematical
Society, Series 2, vol. 42, pp. 230-265, 1936.
[16] M. L. Minsky, and Papert, Seymour S., Perceptrons: An Introduction to Computational Geometry. Cambridge, MA: MIT Press, 1969.
[17] D. E. Rumelhart, and McClelland, J.L. editors, "Parallel Distributed Processing: Explorations in the Microstructures of Cognition."
vol. 1 - Foundations Cambridge, MA: MIT Press, 1986.
[18] J. L. McClelland, and Rumelhart, D.E., "Parallel Distributed Processing: Explorations in the Microstructures of Cognition." vol. 2 Psychological and Biological Models Cambridge, MA: MIT Press, 1986.
[19] D. O. Hebb, The Organization of Behavior. New York: John Wiley & Sons, 1949.
References
[20] W. S. McCulloch, and Pitts, Walter H., "A Logical Calculus of the Ideas Immanent in Neural Nets," Bulletin of Mathematical Biology,
vol. 52, pp. 99 - 115, 1943.
[21] F. Rosenblatt, "The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain," Psychological Review,
vol. 65, pp. 386 - 408, 1958.
[22] F. Rosenblatt, Principles of Neurodynamics. Washington, DC: Spartan Books, 1962.
[23] J. Hawkins, with Sandra Blakeslee, On Intelligence, First ed. New York: Times Books, Henry Holt and Company, 2004.
[24] D. George, and Hawkins, J., "Belief Propagation and Wiring Length Optimization as Organizing Principles for Cortical
Microcircuits," Numenta, Inc., 2005.
[25] D. George, and Hawkins, J., "Invariant Pattern Recognition using Bayesian Inference on Hierarchical Sequences," Numenta, Inc.,
2006.
[26] J. Hawkins, and Dileep George, "Hierarchical Temporal Memory, Concepts, Theory, and Terminology," Numenta, Inc., 2006.
[27] J. Hawkins, "Hierarchical Temporal Memory (HTM): Biological Mapping to Neocortex and Thalamus," Numenta, Inc., 2007.
[28] J. Hawkins, "An Investigation of Adaptive Behavior Towards a Theory of Neocortical Function," 1986.
[29] D. George, "How the Brain Might Work: A Hierarchical and Temporal Model for Learning and Recognition," Doctoral Dissertation;
Stanford University, 2008.