Hacker Timesnew | past | comments | ask | show | jobs | submitlogin

I wonder if the author is referring to the oft-repeated claim that Minsky's and Papert's proof that a perceptron cannot learn the Xor function had a chilling effect on research into neural networks generally, even though Minsky and Papert themselves had shown that multi-layer networks were capable of doing so [1][2].

I realize that even this alleged misunderstanding is not the same as a claim that AI is impossible. The closest attempt of a mathematical proof of the impossibility of AI that I am aware of is the Lucas-Penrose argument from Gödel's first incompleteness theorem [3].

[1] https://en.wikipedia.org/wiki/Perceptrons_(book)

[2] Minsky M. L. and Papert S. A. 1969. Perceptrons. Cambridge, MA: MIT Press.

[3] http://www.iep.utm.edu/lp-argue/



Thanks, the first ref looks like it may be the one.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: