4 ms·
There's an entire category of bit-twiddling methods for different operations. They were developed by programmers at places like the AI labs at MIT and Stanford
by neilv 4y ago
There's an entire category of bit-twiddling methods for different operations.
They were developed by programmers at places like the AI labs at MIT and Stanford, in much earlier days of computing, when size and time constraints sounded much more in everyone's faces than today.
- hgsgm 4y agoIn the old days where programmers knew that everything computers did was AI.
- neilv 4y agoIIUC, Marvin Minsky, at some early point in "AI", wanted to build it with computers as a method for understanding how the human mind works. I think Computer Science came a little later. When I got into "AI", much later, coming from CS and software engineering thinking, "AI" seemed to be "things we currently think are hard for computers to do (and useful computational methods that we've already figured out)". Now "AI" is getting different, and generalized AI is looking more plausibly attainable than it was shortly before DL. (Minsky thought it would happen quicker, and also speculated explosive growth of capability once the AI could teach and evolve itself.)
- TonyTrapp 4y agoRelevant link: https://graphics.stanford.edu/~seander/bithacks.html https://graphics.stanford.edu/~seander/bithacks.html
- delta_p_delta_x 4y agoAlso related: Quake III fast inverse square root[1]. [1]: https://github.com/id-Software/Quake-III-Arena/blob/master/code/game/q_math.c#L552 https://github.com/id-Software/Quake-III-Arena/blob/master/c...
- mkl 4y agoIt's way older than Quake III: https://en.wikipedia.org/wiki/Fast_inverse_square_root https://en.wikipedia.org/wiki/Fast_inverse_square_root Many past discussions on HN: https://hn.algolia.com/?q=inverse+square+root https://hn.algolia.com/?q=inverse+square+root