papers AI Learner
The Github is limit! Click to go to the new site.

Identity Crisis: Memorization and Generalization under Extreme Overparameterization

2019-02-13
Chiyuan Zhang, Samy Bengio, Moritz Hardt, Yoram Singer

Abstract

We study the interplay between memorization and generalization of overparametrized networks in the extreme case of a single training example. The learning task is to predict an output which is as similar as possible to the input. We examine both fully-connected and convolutional networks that are initialized randomly and then trained to minimize the reconstruction error. The trained networks take one of the two forms: the constant function (“memorization”) and the identity function (“generalization”). We show that different architectures exhibit vastly different inductive bias towards memorization and generalization. An important consequence of our study is that even in extreme cases of overparameterization, deep learning can result in proper generalization.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1902.04698

PDF

http://arxiv.org/pdf/1902.04698


Similar Posts

Comments