About this time last year I had an optimization problem involving linear combinations of Voronoi cell centroids and I basically used random search (I can't remember why differential evolution didn't work).
I don't know much, say, Keras, but I think I understand how to implement backprop-like behaviors with Autograd. Maybe even sandwich concept-specific models between fully-connected linear+sigmoid layers to give them some oomph.
Comments
Interesting.
About this time last year I had an optimization problem involving linear combinations of Voronoi cell centroids and I basically used random search (I can't remember why differential evolution didn't work).
I don't know much, say, Keras, but I think I understand how to implement backprop-like behaviors with Autograd. Maybe even sandwich concept-specific models between fully-connected linear+sigmoid layers to give them some oomph.