Alan Turing on Learning Machines

We cannot expect to find a good child-machine at the first attempt. One must experiment with teaching one such machine and see how well it learns. One can then try another and see if it is better or worse. There is an obvious connection between this process and evolution, by the identifications
  • Structure of the child-machine = Hereditary material
  • Changes of the child-machine = Mutations
  • Natural selection = Judgement of the experimentor
One may hope, however, that this process will be more expeditious than evolution. The survival of the fittest is a slow method for measuring advantages. The experimenter, by the exercise of intelligence, should be able to speed it up. Equally important is the fact that he is not restricted to random mutations. If he can trace a cause for some weakness he can probably think of the kind of mutation which will improve it.


Folksonomies: new media information technology

/science/medicine/genetics (0.568016)
/style and fashion (0.510550)
/technology and computing/computer certification (0.464701)

li>Structure of the child-machine = Hereditary material (0.952478 (negative:-0.405875)), random mutations (0.889557 (neutral:0.000000)), obvious connection (0.884134 (neutral:0.000000)), Hereditary material (0.855842 (negative:-0.405875)), good child-machine (0.844428 (negative:-0.609663)), slow method (0.804299 (positive:0.447077)), Natural selection (0.792725 (neutral:0.000000)), process (0.440689 (neutral:0.000000)), evolution (0.440125 (neutral:0.000000)), identifications (0.396407 (neutral:0.000000)), experimenter (0.387006 (neutral:0.000000)), Judgement (0.373740 (neutral:0.000000)), weakness (0.373672 (negative:-0.291758)), attempt (0.365152 (negative:-0.609663)), advantages (0.337289 (positive:0.447077)), exercise (0.328266 (neutral:0.000000)), kind (0.328090 (negative:-0.291758)), survival (0.326702 (positive:0.447077)), cause (0.322005 (negative:-0.291758)), Changes (0.316868 (neutral:0.000000)), fact (0.314760 (neutral:0.000000)), experimentor (0.313557 (neutral:0.000000))

Natural selection (0.975266): dbpedia | freebase
Mutation (0.846299): dbpedia | freebase
Evolution (0.765934): dbpedia | freebase | opencyc
Genetics (0.689633): dbpedia | freebase | opencyc
Experiment (0.614226): dbpedia | freebase | opencyc
DNA repair (0.573997): dbpedia | freebase | yago
DNA (0.554107): website | dbpedia | freebase | yago
Evolutionary biology (0.528515): dbpedia | freebase | opencyc

 Computing Machinery and Intelligence
Periodicals>Journal Article:  Turing, Alan (October 1950), Computing Machinery and Intelligence, Mind: A Quarterly Review of Psychology and Philosophy, 59(236):433-460, Retrieved on -0001-11-30