On Robustness and Transferability of Convolutional Neural Networks

07/16/2020 ∙ by Josip Djolonga, et al. ∙ 15

Modern deep convolutional networks (CNNs) are often criticized for not generalizing under distributional shifts. However, several recent breakthroughs in transfer learning suggest that these networks can cope with severe distribution shifts and successfully adapt to new tasks from a few training examples. In this work we revisit the out-of-distribution and transfer performance of modern image classification CNNs and investigate the impact of the pre-training data size, the model scale, and the data preprocessing pipeline. We find that increasing both the training set and model sizes significantly improve the distributional shift robustness. Furthermore, we show that, perhaps surprisingly, simple changes in the preprocessing such as modifying the image resolution can significantly mitigate robustness issues in some cases. Finally, we outline the shortcomings of existing robustness evaluation datasets and introduce a synthetic dataset we use for a systematic analysis across common factors of variation.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 5

page 7

page 15

page 16

page 18

page 20

page 21

page 22

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.