Code for our paper on best practices to train neural networks with direct feedback alignment (DFA).
Primary LanguagePython