Skip to content

Comment on JAX: Numpy with Gradients, GPUs and TPUsparent

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.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.