# Cost-Sensitive Robustness against Adversarial Examples

arXiv: https://arxiv.org/abs/1810.09225 · HTML: https://mlsanity.com/paper/1810.09225/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Oct 2018.

## Built on by (most cited first, 9 of 9)

- [Advances in adversarial attacks and defenses in computer vision: A survey](https://mlsanity.com/paper/2108.00401/) (2021)
- [Auditing Membership Leakages of Multi-Exit Networks](https://mlsanity.com/paper/2208.11180/) (2022)
- [Adversarial Robustness for Tabular Data through Cost and Utility Awareness](https://mlsanity.com/paper/2208.13058/) (2022)
- [Adversarial Learning with Cost-Sensitive Classes](https://mlsanity.com/paper/2101.12372/) (2021)
- [Robustness Testing of Data and Knowledge Driven Anomaly Detection in Cyber-Physical Systems](https://mlsanity.com/paper/2204.09183/) (2022)
- [Transferable Adversarial Robustness for Categorical Data via Universal Robust Embeddings](https://mlsanity.com/paper/2306.04064/) (2023)
- [Rethinking Cost-sensitive Classification in Deep Learning via Adversarial Data Augmentation](https://mlsanity.com/paper/2208.11739/) (2022)
- [Context-Adaptive Deep Neural Networks via Bridge-Mode Connectivity](https://mlsanity.com/paper/2211.15436/) (2022)
- [Provably Cost-Sensitive Adversarial Defense via Randomized Smoothing](https://mlsanity.com/paper/2310.08732/) (2023)

## Abstract (arXiv)

Several recent works have developed methods for training classifiers that are certifiably robust against norm-bounded adversarial perturbations. These methods assume that all the adversarial transformations are equally important, which is seldom the case in real-world applications. We advocate for cost-sensitive robustness as the criteria for measuring the classifier's performance for tasks where some adversarial transformation are more important than others. We encode the potential harm of each adversarial transformation in a cost matrix, and propose a general objective function to adapt the robust training method of Wong & Kolter (2018) to optimize for cost-sensitive robustness. Our experiments on simple MNIST and CIFAR10 models with a variety of cost matrices show that the proposed approach can produce models with substantially reduced cost-sensitive robust error, while maintaining classification accuracy.