pith. sign in

arxiv: 2106.10989 · v4 · pith:XYPXL5L7new · submitted 2021-06-21 · 💻 cs.CV · cs.AI

ImageNet Pre-training also Transfers Non-Robustness

classification 💻 cs.CV cs.AI
keywords modelimagenetnon-robustnesspre-trainingfine-tunedpre-trainedadversarialconducted
0
0 comments X
read the original abstract

ImageNet pre-training has enabled state-of-the-art results on many tasks. In spite of its recognized contribution to generalization, we observed in this study that ImageNet pre-training also transfers adversarial non-robustness from pre-trained model into fine-tuned model in the downstream classification tasks. We first conducted experiments on various datasets and network backbones to uncover the adversarial non-robustness in fine-tuned model. Further analysis was conducted on examining the learned knowledge of fine-tuned model and standard model, and revealed that the reason leading to the non-robustness is the non-robust features transferred from ImageNet pre-trained model. Finally, we analyzed the preference for feature learning of the pre-trained model, explored the factors influencing robustness, and introduced a simple robust ImageNet pre-training solution. Our code is available at \url{https://github.com/jiamingzhang94/ImageNet-Pretraining-transfers-non-robustness}.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.