A Majority Invariant Approach to Patch Robustness Certification for Deep Learning Models

2023/08/01 by Qilin Zhou, Zhengyuan Wei, Zhou, Qilin +5 · 1 citation
Computer Science · #Adversarial Robustness in Machine Learning #Computer Vision and Pattern Recognition (cs.CV) #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning and Data Classification #Software Engineering (cs.SE)

paper · pdf · doi:10.48550/arxiv.2308.00452

openalex publication_date 2023/08/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Patch robustness certification ensures no patch within a given bound on a sample can manipulate a deep learning model to predict a different label. However, existing techniques cannot certify samples that cannot meet their strict bars at the classifier or patch region levels. This paper proposes MajorCert. MajorCert firstly finds all possible label sets manipulatable by the same patch region on the same sample across the underlying classifiers, then enumerates their combinations element-wise, and finally checks whether the majority invariant of all these combinations is intact to certify samples.

Cited by

• Scalable and Precise Patch Robustness Certification for Deep Learning Models with Top-k Predictions

Related

• CrossCert: A Cross-Checking Detection Approach to Patch Robustness Certification for Deep Learning Models
• Scalable and Precise Patch Robustness Certification for Deep Learning Models with Top-k Predictions
• Toward Patch Robustness Certification and Detection for Deep Learning Systems Beyond Consistent Samples
• PatchDEMUX: A Certifiably Robust Framework for Multi-label Classifiers Against Adversarial Patches
• Enhancing Certifiable Robustness via a Deep Model Ensemble