This year the workshop aims to bring together young researchers (PhD students, postdocs, and young faculty members) working (or interested) in the development of a better understanding of deep learning, particularly the interplay between the statistical and optimization aspects. It will feature tutorial talks by Peter Bartlett (UC Berkeley), Nathan Srebro (TTI Chicago), Jason Lee (Marshall School of Business) and Johannes Schmidt-Hieber (University of Twente), among other invited and contributed talks.