mlsanityData to 7 Oct 2026

Towards Consistency in Adversarial Classification

arXiv 2205.10022 ↗First posted 20 May 2022cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

In this paper, we study the problem of consistency in the context of adversarial examples. Specifically, we tackle the following question: can surrogate losses still be used as a proxy for minimizing the 0/1 loss in the presence of an adversary that alters the inputs at test-time? Different from the standard classification task, this question cannot be reduced to a point-wise minimization problem, and calibration needs not to be sufficient to ensure consistency. In this paper, we expose some pathological behaviors specific to the adversarial problem, and show that no convex surrogate loss can be consistent or calibrated in this context. It is therefore necessary to design another class of surrogate functions that can be used to solve the adversarial consistency issue. As a first step towards designing such a class, we identify sufficient and necessary conditions for a surrogate loss to be calibrated in both the adversarial and standard settings. Finally, we give some directions for building a class of losses that could be consistent in the adversarial framework.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 14 of its 22 references that are in the map. First posted on arXiv 20 May 2022.

Builds on

14 of 22 references.

References · 14

  1. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  2. 2013Intriguing properties of neural networkscited by 5,114
  3. 2017Evasion Attacks against Machine Learning at Test Timecited by 686
  4. 2020An Analysis of Adversarial Attacks and Defenses on Autonomous Driving Modelscited by 67
  5. 2019Lower Bounds on Adversarial Robustness from Optimal Transportcited by 52
  6. 2021Calibration and Consistency of Adversarial Surrogate Lossescited by 47
  7. 2020Randomization matters. How to defend against strong adversarial attackscited by 37
  8. 2019Adversarial Risk via Optimal Transport and Optimal Couplingscited by 36
  9. 2020Calibrated Surrogate Losses for Adversarially Robust Classificationcited by 36
  10. 2022The Many Faces of Adversarial Riskcited by 28
  11. 2021A Finer Calibration Analysis for Adversarial Robustnesscited by 26
  12. 2021On the Existence of the Adversarial Bayes Classifier (Extended Version)cited by 25
2 more
  1. 2021Mixed Nash Equilibria in the Adversarial Examples Gamecited by 22
  2. 2021The Geometry of Adversarial Training in Binary Classificationcited by 20

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 4; 2024: 3; 2025: 3.

12 papers.

  1. 2023Cross-Entropy Loss Functions: Theoretical Analysis and Applicationscited by 206
  2. 2025Adversarial Robustness in Two-Stage Learning-to-Defer: Algorithms and Guaranteescited by 16
  3. 2022Existence and Minimax Theorems for Adversarial Surrogate Risks in Binary Classificationcited by 15
  4. 2023Adversarial Training Should Be Cast as a Non-Zero-Sum Gamecited by 13
  5. 2023The Adversarial Consistency of Surrogate Risks for Binary Classificationcited by 6
  6. 2022The Consistency of Adversarial Training for Binary Classificationcited by 3
  7. 2023Non-Asymptotic Bounds for Adversarial Excess Risk under Misspecified Modelscited by 2
  8. 2024Adversarial Consistency and the Uniqueness of the Adversarial Bayes Classifiercited by 2
  9. 2024Uniform Convergence of Adversarially Robust Classifierscited by 1
  10. 2025Adversarial Surrogate Risk Bounds for Binary Classificationcited by 1
  11. 2024Towards Calibrated Losses for Adversarial Robust Reject Option Classificationcited by 0
  12. 2025Unveiling the Role of Randomization in Multiclass Adversarial Classification: Insights from Graph Theorycited by 0