NIPS Proceedingsβ

The Implicit Bias of AdaGrad on Separable Data

Part of: Advances in Neural Information Processing Systems 32 (NIPS 2019)

[PDF] [BibTeX] [Supplemental] [Reviews] [Author Feedback] [Meta Review] [Sourcecode]


Conference Event Type: Poster


We study the implicit bias of AdaGrad on separable linear classification problems. We show that AdaGrad converges to a direction that can be characterized as the solution of a quadratic optimization problem with the same feasible set as the hard SVM problem. We also give a discussion about how different choices of the hyperparameters of AdaGrad may impact this direction. This provides a deeper understanding of why adaptive methods do not seem to have the generalization ability as good as gradient descent does in practice.