Generalization in Deep Learning (arXiv)

This year's ICLR 2017 best paper Understanding deep learning requires rethinking generalization reinvigorated interest in gaining a better understanding of the generalization behaviour of deep neural networks. In this tradition, this paper with Yoshua Bengio as co-author seeks to provide new theoretical explanations and new direct analyses for generalization in Deep Learning. It also proposes a new family lf generalization terms that takes these new insights into account.


Want to receive more content like this in your inbox?