Brandon Amos

I've not run into this before but do let me know if you find an easy fix I should put back into the repo > On Sep 9, 2019, at...

I think this is the same as Eq (3) of https://arxiv.org/abs/1802.08665, right?  If so, it would be cool to mention the connection to that paper, and to https://arxiv.org/abs/1106.1925, since...

Hi, that was an external contribution coming from #26 and I didn't test that code. Let me know if you find any other details out. If it's not working I'd...

I'm also now hitting a bottleneck in serializing the canonicalization derivatives and am adding in similar batching here. > the changes [in this PR](https://github.com/cvxgrp/cvxpy/pull/1010) dramatically speed up the creation of...

> Maybe we could expand apply_parameters to take in a matrix of parameters (one column for each item in the batch)? @bamos is this what you mean by batching? Yeah,...

Hmm, would it be easy to add something like `split_dual` and `split_dual_adjoint` to `ParamConeProg` in CVXPY using most of the existing code here and then expose that in `CvxpyLayer` here...

(Sorry accidentally pressed a hotkey and sent that in early, edited)

Ah thanks -- that's pretty hidden and also doesn't hold in my case where I replaced that with my own recurrent policy class. Maybe there should be a well-placed comment...

Hi, thanks for taking a closer look! For better intuition on all of that, I recommend considering a quadratic function (e.g. y=x^T Q x) instead of the optimization problem (y=argmin...

That's correct, we don't have general state constraints implemented