Parametric Study of a Genetic Algorithm (2003) [pdf]

Re:Parametric Study of a Genetic Algorithm (2003) [pdf] by TezlaOil on Sunday August 06, 2017 @02:20PM (#14942586)

Machine learning is automated classification, regression, and decision-making. Genetic algorithms don't tend to perform so well in these areas (just as ML is not so appropriate for combinatorial optimization).

Instead, genetic methods work well for various messy industrial engineering such as strip packaging design or antenna design. One of my favorite recent results solves a structural engineering problem using grammatical evolution [1].

[1] http://www.sciencedirect.com/science/article/pii/S0926580513...

[ Parent | on HN ]

Re:Parametric Study of a Genetic Algorithm (2003) [pdf] by amelius on Sunday August 06, 2017 @02:43PM (#14942730)

Perhaps genetic algorithms could be useful for hyperparameter optimization of deep learning methods? ;)

[ Parent | on HN ]

Re:Parametric Study of a Genetic Algorithm (2003) [pdf] by thearn4 on Monday August 07, 2017 @02:12PM (#14949756)

I remember using GAs to tune SVM kernel hyperparameters a few years ago. They worked well enough, but I mostly did it because I was lazy.

[ Parent | on HN ]

Re:Parametric Study of a Genetic Algorithm (2003) [pdf] by R_haterade on Sunday August 06, 2017 @11:39PM (#14945155)

It's been looked into. I'm really struggling to remember who was doing the looking though, but it exists.

[ Parent | on HN ]

Re:Parametric Study of a Genetic Algorithm (2003) [pdf] by bstamour on Sunday August 06, 2017 @10:13PM (#14944843)

If you have the horsepower to test a population of deep nets, then absolutely.

[ Parent | on HN ]