Natural gradients for deep layers

Open parvkpr opened this issue 5 years ago • 1 comments

In NES algorithms do we use backpropogation from the last layers gradients(computed by the objectve function). I am curious as to how to optimize the hidden layers since they are not directly affecting the objective function

Feb 06 '20 18:02 parvkpr

Evolution strategies do not use backpropagation, but they differentiate (approximately) across a population of solutions, rather than across parameters within a solution. Perhaps reading this blog post is useful :stuck_out_tongue:

Apr 06 '20 13:04 jinyeom