News for hackers. Stuff that matters.
Re:Machine Learning by dmitriy_ko on Wednesday May 17, 2017 @04:42AM (#14356621)
With gradient descent you are walking in the direction of negative gradient. Stirring the pile implies that you are walking randomly from point to point.
Re:Machine Learning by TeMPOraL on Wednesday May 17, 2017 @05:43AM (#14356905)
No, by stirring you're just making the pile to move, and things descend down the gradient of gravity. :).
Re:Machine Learning by steanne on Wednesday May 17, 2017 @10:40AM (#14358850)
The little things moreso. Brazil nut effect!
Re:Machine Learning by yakult on Wednesday May 17, 2017 @05:51AM (#14356942)
When you stir, you're stirring in a pattern, probably a circular one. Plus you're probably not paying too much attention to the shape of the pile, so it's gradient-free. It's pattern search. Except it's only approximately a circular pattern. There are random variations. Stochastic pattern search.
Re:Machine Learning by andai on Wednesday May 17, 2017 @05:11AM (#14356760)
So more like a genetic algorithm then?
Re:Machine Learning by randomsearch on Wednesday May 17, 2017 @02:06PM (#14361055)
GAs also exploit structure in the search space.

To see this most clearly, read up on estimation of distribution algorithms (EDAs), which generalise the ways in which GAs work.