leogao comments on Gradient descent is not just more efficient genetic algorithms

leogao 10 Sep 2021 23:55 UTC
2 points
It’s relevant because it demonstrates that in differentiable functions, if it is the case that changing only $w_{1}$ does not affect the loss, and changing only $w_{2}$ does not affect the loss, then it is not possible that changing them both can affect the loss either.