Talk
Deep Learning through the Lens of Data
- Gintare Karolina Dziugaite (Google Brain)
Abstract
Deep learning comes with excessive demands for data. In this talk, I will present my recent work on showing that not all data is necessary for training an accurate predictor. In particular, one can drop “easy-to-learn” examples, and do just as well as learning on all of the data. Given this disparate “importance” of training data on generalization, I will present empirical analysis of the loss landscape derived from different subsets of the training examples. I will then look into how the training dynamics are influenced by easy versus hard data.