Algorithmic Regularization in Over-parameterized Matrix Recovery and Neural Networks with Quadratic Activations

Neyman Seminar
Feb 28, 2018 4:00pm to 5:00pm
1011 Evans Hall
Happening As Scheduled
Over-parameterized models are widely and successfully used in deep learning, but their workings are far from understood. In many practical scenarios, the learned model generalizes to the test data, even though the hypothesis class contains a model that completely overfits the training data and no regularization is applied. In this talk, we will show that such phenomenon occurs in...
Tengyu Ma, Facebook AI Research