Some time ago, I shared a starter kit for reinforcement learning research with Ray and RLlib. The following is a productive tech stack for doing deep learning research.
I have published a paper version of my patent on Arxiv, titled “Iterative Training: Finding Binary Weight Deep Neural Networks with Layer Binarization”.
This is a guide for increasing GPU training throughput given a batch of training run, as opposed to increasing instantaneous throughput for one training run. Obviously, doing the latter can increase training throughput, but that is not in the scope of this guide.