Re:Evolution Strategies as a Scalable Alternative to Reinfor...
by mark_l_watson on Saturday March 25, 2017 @11:32AM
(#13956104)
Great write up.
I had a similar example in my 20 year old book 'C++ Power Paradigms' in which I used a genetic algorithm to train the weights in a recurrent network. As a performance hack, weights were initially represented by just a few bits, and the bit length would gradually be increased, which greatly increased the search space. I never got this to scale past small networks, but I have thought of revisiting my old code since I have a lot more computing power available now, compared to 20 years ago.
I had a similar example in my 20 year old book 'C++ Power Paradigms' in which I used a genetic algorithm to train the weights in a recurrent network. As a performance hack, weights were initially represented by just a few bits, and the bit length would gradually be increased, which greatly increased the search space. I never got this to scale past small networks, but I have thought of revisiting my old code since I have a lot more computing power available now, compared to 20 years ago.