Computer Chess History by Bill WallA timeline of computer chess, starting in 1947. There’s nothing particularly about learning.
A few landmarks in machine learning history. They are now far from the main streams of research, but it is good for us to remember where we have rowed.
| Jay : game learning : history |
Computer Chess History by Bill Wall
A timeline of computer chess, starting in 1947. There’s nothing particularly about learning.
Some Studies in Machine Learning Using the Game of Checkers II—Recent Progress by
Arthur Samuel (1967 PDF)
The second of two papers about the great original work, far ahead of its time. (The first paper was 1959.) This makes fascinating reading.
Samuel’s Checkers Player (1998)
From the book Reinforcement Learning: An Introduction by Richard S. Sutton and Andrew G. Barto (a good book, by the way).
This case-study compares Samuel’s original learning methods with modern temporal-difference learning.
Matchbox tic-tac-toe (early 1960’s)
Donald Michie
Michie designed MENACE, the Matchbox Educable Noughts And Crosses Engine, around 1960. It’s a pioneering experiment in machine learning, although it doesn’t involve a machine! It’s still a good introductory example for students.
This mention is from a hobbyist magazine in 1984.
Deep Thought’s evaluation function tuner (1988)
Andreas Nowatzyk
Source code from 1988 to tune the evaluator of Deep Thought, the chess machine
predecessor of Deep Blue. The code examines grandmaster games and adjusts
evaluation parameters.
Warehoused by Tim Mann.