# Local Reweighting for Adversarial Training

arXiv: https://arxiv.org/abs/2106.15776 · HTML: https://mlsanity.com/paper/2106.15776/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 40 references that are in the map.
- First posted on arXiv 30 Jun 2021.

## Builds on (most cited first, 28 of 28)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [Towards Evaluating the Robustness of Neural Networks](https://mlsanity.com/paper/1608.04644/) (2016)
- [Adversarial examples in the physical world](https://mlsanity.com/paper/1607.02533/) (2016)
- [Theoretically Principled Trade-off between Robustness and Accuracy](https://mlsanity.com/paper/1901.08573/) (2019)
- [Certified Adversarial Robustness via Randomized Smoothing](https://mlsanity.com/paper/1902.02918/) (2019)
- [Robustness May Be at Odds with Accuracy](https://mlsanity.com/paper/1805.12152/) (2018)
- [Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images](https://mlsanity.com/paper/1412.1897/) (2014)
- [Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methods](https://mlsanity.com/paper/1705.07263/) (2017)
- [Overfitting in adversarially robust deep learning](https://mlsanity.com/paper/2002.11569/) (2020)
- [On Evaluating Adversarial Robustness](https://mlsanity.com/paper/1902.06705/) (2019)
- [Adversarial Training Methods for Semi-Supervised Text Classification](https://mlsanity.com/paper/1605.07725/) (2016)
- [Attacks Which Do Not Kill Training Make Adversarial Learning Stronger](https://mlsanity.com/paper/2002.11242/) (2020)
- [DeepDriving: Learning Affordance for Direct Perception in Autonomous Driving](https://mlsanity.com/paper/1505.00256/) (2015)
- [Improving Adversarial Robustness via Promoting Ensemble Diversity](https://mlsanity.com/paper/1901.08846/) (2019)
- [Geometry-aware Instance-reweighted Adversarial Training](https://mlsanity.com/paper/2010.01736/) (2020)
- [Understanding Adversarial Attacks on Deep Learning Based Medical Image Analysis Systems](https://mlsanity.com/paper/1907.10456/) (2019)
- [Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuning](https://mlsanity.com/paper/2003.12862/) (2020)
- [Understanding and Mitigating the Tradeoff Between Robustness and Accuracy](https://mlsanity.com/paper/2002.10716/) (2020)
- [Towards the Science of Security and Privacy in Machine Learning](https://mlsanity.com/paper/1611.03814/) (2016)
- [Curriculum Adversarial Training](https://mlsanity.com/paper/1805.04807/) (2018)
- [Universal Adversarial Training](https://mlsanity.com/paper/1811.11304/) (2018)
- [Once-for-All Adversarial Training: In-Situ Tradeoff between Robustness and Accuracy for Free](https://mlsanity.com/paper/2010.11828/) (2020)
- [Understanding the Interaction of Adversarial Training with Noisy Labels](https://mlsanity.com/paper/2102.03482/) (2021)
- [Learning Diverse-Structured Networks for Adversarial Robustness](https://mlsanity.com/paper/2102.01886/) (2021)

## Built on by (most cited first, 1 of 1)

- [Latent Feature Relation Consistency for Adversarial Robustness](https://mlsanity.com/paper/2303.16697/) (2023)

## Abstract (arXiv)

Instances-reweighted adversarial training (IRAT) can significantly boost the robustness of trained models, where data being less/more vulnerable to the given attack are assigned smaller/larger weights during training. However, when tested on attacks different from the given attack simulated in training, the robustness may drop significantly (e.g., even worse than no reweighting). In this paper, we study this problem and propose our solution--locally reweighted adversarial training (LRAT). The rationale behind IRAT is that we do not need to pay much attention to an instance that is already safe under the attack. We argue that the safeness should be attack-dependent, so that for the same instance, its weight can change given different attacks based on the same model. Thus, if the attack simulated in training is mis-specified, the weights of IRAT are misleading. To this end, LRAT pairs each instance with its adversarial variants and performs local reweighting inside each pair, while performing no global reweighting--the rationale is to fit the instance itself if it is immune to the attack, but not to skip the pair, in order to passively defend different attacks in future. Experiments show that LRAT works better than both IRAT (i.e., global reweighting) and the standard AT (i.e., no reweighting) when trained with an attack and tested on different attacks.