# Reinforcing Adversarial Robustness using Model Confidence Induced by Adversarial Training

arXiv: https://arxiv.org/abs/1711.08001 · HTML: https://mlsanity.com/paper/1711.08001/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Nov 2017.

## Built on by (most cited first, 4 of 4)

- [Relating Adversarially Robust Generalization to Flat Minima](https://mlsanity.com/paper/2104.04448/) (2021)
- [Two Coupled Rejection Metrics Can Tell Adversarial Examples Apart](https://mlsanity.com/paper/2105.14785/) (2021)
- [GenCode: A Generic Data Augmentation Framework for Boosting Deep Learning-Based Code Understanding](https://mlsanity.com/paper/2402.15769/) (2024)
- [Robust Unsupervised Domain Adaptation for 3D Point Cloud Segmentation Under Source Adversarial Attacks](https://mlsanity.com/paper/2504.01659/) (2025)

## Abstract (arXiv)

In this paper we study leveraging confidence information induced by adversarial training to reinforce adversarial robustness of a given adversarially trained model. A natural measure of confidence is |F(bf x)|_∞ (i.e. how confident F is about its prediction?). We start by analyzing an adversarial training formulation proposed by Madry et al.. We demonstrate that, under a variety of instantiations, an only somewhat good solution to their objective induces confidence to be a discriminator, which can distinguish between right and wrong model predictions in a neighborhood of a point sampled from the underlying distribution. Based on this, we propose Highly Confident Near Neighbor (tt HCNN), a framework that combines confidence information and nearest neighbor search, to reinforce adversarial robustness of a base model. We give algorithms in this framework and perform a detailed empirical study. We report encouraging experimental results that support our analysis, and also discuss problems we observed with existing adversarial training.