Transcript René Descartes (1596

Artificial Intelligence
René Descartes
(1596-1650)
hydraulic theory
Joseph Jaquard
1752-1834
Loom
Punch-card
program
Product
Joseph Jaquard
Charles Babbage
(1791-1871)
‘difference engine’
Herman Hollerith
1860-1929
coding
The pins that were positioned over
holes would be able to pass through the
cards. Upon passing through the cards,
the pins became submerged in mercury,
thus completing the circuit.
The Hollerith Cards, as they became
known, could be prepared in any
order and stored in a logical manner.
Alan Turing
(1912-1954)
German
Enigma
machine
Relay-based
computer
Computer-based Artificial Intelligence
Checkers too simple?
Chess too simple?
Kasparov
vs. Big Blue
Perceptron
1928 - 1969
The perceptron is a type of artificial neural network invented
in 1957 at Cornell by Frank Rosenblatt. It can be seen as the
simplest kind of feedforward neural network: a linear classifier.
"the fundamental laws of organization which are common to all information
handling systems, machines and men included, may eventually be understood."
Parallel Distributed Processing
A multilayer perceptron (PDP network).
The network ‘learns’ weights through examples
(learns distances between cities in Alberta).