paper-with-me

홈 › Papers

When Robustness Doesn’t Promote Robustness: Synthetic vs. Natural Distribution Shifts on ImageNet

2019-09-25 · Rohan Taori, Achal Dave, Vaishaal Shankar, Nicholas Carlini, Benjamin Recht, Ludwig Schmidt

We conduct a large experimental comparison of various robustness metrics for image classification. The main question of our study is to what extent current synthetic robustness interventions (lp-adversarial examples, noise corruptions, etc.) promote robustness under natural distribution shifts occurring in real data. To this end, we evaluate 147 ImageNet models under 199 different evaluation settings. We find that no current robustness intervention improves robustness on natural distribution shifts beyond a baseline given by standard models without a robustness intervention. The only exception is the use of larger training datasets, which provides a small increase in robustness on one natural distribution shift. Our results indicate that robustness improvements on real data may require new methodology and more evaluations on natural distribution shifts.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

image-classificationImage Classification

Similar Papers 제목 키워드 기반

Adversarial Vulnerability as a Consequence of On-Manifold Inseparibility

2024-10-09 · Rajdeep Haldar, Yue Xing, Qifan Song, Guang Lin

Recent works have shown theoretically and empirically that redundant data dimensions are a source of adversarial vulnerability. However, the inverse doesn't seem to hold in practice; employing dimension-reduction techniq…

AttributeDimensionality ReductionSecond-order methods

Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness

2023-05-08 · Liangliang Cao, BoWen Zhang, Chen Chen, Yinfei Yang 외

The CLIP (Contrastive Language-Image Pre-training) model and its variants are becoming the de facto backbone in many applications. However, training a CLIP model from hundreds of millions of image-text pairs can be prohi…

Adversarial TextRetrieval

Extending Kernel PCA through Dualization: Sparsity, Robustness and Fast Algorithms

2023-06-09 · Francesco Tonin, Alex Lambert, Panagiotis Patrinos, Johan A. K. Suykens

The goal of this paper is to revisit Kernel Principal Component Analysis (KPCA) through dualization of a difference of convex functions. This allows to naturally extend KPCA to multiple objective functions and leads to e…

On the Effect of Low-Rank Weights on Adversarial Robustness of Neural Networks

2019-01-29 · Peter Langenberg, Emilio Rafael Balda, Arash Behboodi, Rudolf Mathar

Recently, there has been an abundance of works on designing Deep Neural Networks (DNNs) that are robust to adversarial examples. In particular, a central question is which features of DNNs influence adversarial robustnes…

Adversarial Robustness

Using Synthetic Corruptions to Measure Robustness to Natural Distribution Shifts

2021-07-26 · Alfred Laugros, Alice Caplier, Matthieu Ospici

Synthetic corruptions gathered into a benchmark are frequently used to measure neural network robustness to distribution shifts. However, robustness to synthetic corruption benchmarks is not always predictive of robustne…