1846
Comments (69)
sorted by:
You're viewing a single comment thread. View all comments, or full comment thread.
31
Zskills 31 points ago +31 / -0

I'm a software engineer, took a year of discrete mathematics (combinatorics, graph theory etc), year of linear algebra, calculus through dif. EQ, studied algorithms... statistical analysis, know relevant languages for data analysis like SQL & Python, etc... but I don't know anything about these "genetic" algorithms you are talking about. I do have a degree in human physiology as well, and took a couple years of biology so I understand how genes replicate from the ground up.. but I am not seeing the connection!

18
zakat 18 points ago +18 / -0

Genetic programming is a method to optimize non-linear problems.

You have a set of starting values. You have a fitness function that you can plug the values in.
You can let starting values "mate" with each other and let them basically produce "children". Kids with bad/worse fitness are ignored while you focus on the fit ones.
This reduces the search space since full depth/breadth first search is not feasible.

The process of mating can be quite easy (just take the average of both parents) or complicated. You definitely have the skill to work on the problems.