Changes

Jump to: navigation, search

Donald Michie

154 bytes added, 07:05, 3 June 2021
no edit summary
<span id="MENACE"></span>
=Machine Learning=
Michie began his first experiments in [[Learning|machine learning]] in 1960. His [https://en.wikipedia.org/wiki/Tic-tac-toe tic-tac-toe] machine [https://en.wikipedia.org/wiki/Matchbox_Educable_Noughts_and_Crosses_Engine MENACE ] (Machine Educable Noughts And Crosses Engine) demonstrated the basic principle of a [[Reinforcement Learning|self-reinforcing learning]] mechanism. MENACE employed Michie's conceptually simple general-purpose learning algorithm BOXES <ref>[[Donald Michie]] ('''1961'''). ''Trial and Error''. Penguin Science Survey</ref> <ref> [[Martin Gardner]] ('''1969, 1991'''). ''The Unexpected Hanging and Other Mathematical Diversions''. [https://en.wikipedia.org/wiki/Simon_%26_Schuster Simon & Schuster], [https://en.wikipedia.org/wiki/University_of_Chicago_Press University Of Chicago Press], Chapter 8: A Matchbox Game-Learning Machine</ref> <ref>[[Donald Michie]], Roger A. Chambers ('''1968'''). ''Boxes: An experiment on adaptive control''. In E. Dale and D. Michie, editors, [http://www.doc.ic.ac.uk/%7Eshm/MI/mi2.html Machine Intelligence 2], Edinburgh: Oliver & Boyd, pp. 137-152</ref> <ref>[[Alex Bell]] ('''1972'''). ''[http://www.chilton-computing.org.uk/acl/literature/books/gamesplaying/overview.htm Games Playing with Computers]''. ''[http://www.chilton-computing.org.uk/acl/literature/books/gamesplaying/p001.htm 1.3 NOUGHTS AND CROSSES]'', [https://en.wikipedia.org/wiki/Allen_%26_Unwin Allen & Unwin]</ref> <ref>[http://www.it.uu.se/edu/course/homepage/ai/menace UU/IT/AI exercise: Implementing MENACE], [https://en.wikipedia.org/wiki/Uppsala_University Uppsala University]</ref> which could also discover robust control strategies for the pole balancing problem <ref>[http://pages.cs.wisc.edu/%7Efinton/poledriver.html Controller-less Driver For the Cart-Pole Problem]</ref> <ref>[http://pages.cs.wisc.edu/%7Efinton/qcontroller.html Q-Learning Controller for the Cart-Pole Problem]</ref> , but was soon employed industrially to evolve strategies for automatic control, such as controlling a steel mill <ref>[http://www.aiai.ed.ac.uk/%7Edm/dmcv.html D. Michie CV]</ref>.
[[FILE:jamesbridle-playful.023.jpg|none|border|text-bottom|link=http://shorttermmemoryloss.com/menace/]]
Self build [https://en.wikipedia.org/wiki/Matchbox_Educable_Noughts_and_Crosses_Engine MENACE ] by James Bridle <ref>[http://shorttermmemoryloss.com/menace/ A New Theory of Awesomeness and Miracles] by [http://shorttermmemoryloss.com/ James Bridle]</ref><br/>
in the traditions of [https://en.wikipedia.org/wiki/Heath_Robinson_%28codebreaking_machine%29 Heath Robinson] and [[Mathematician#Babbage|Charles Babbage]] <ref>[http://www.flickr.com/photos/stml/sets/72157622716067016/ Menace - Flickr Photo stream]</ref> <ref>[http://boingboing.net/2009/11/02/mechanical-computer.html Mechanical computer uses matchboxes and beans to learn Tic-Tac-Toe - Boing Boing]</ref>

Navigation menu