Robustness Certification

Robustness certification aims to provide mathematically provable guarantees that a machine learning model's predictions will remain reliable even when its inputs are perturbed, addressing vulnerabilities to adversarial attacks and noise. Current research focuses on developing and improving certification techniques for various model architectures, including convolutional neural networks, graph convolutional networks, vision transformers, and Bayesian neural networks, often employing methods like randomized smoothing and polyhedral analysis. This field is crucial for deploying machine learning models in safety-critical applications, such as autonomous vehicles and medical diagnosis, where reliable predictions under uncertainty are paramount.

Papers

August 2, 2024

Certifiably Robust Encoding Schemes
Aman Saxena, Tom Wollschläger, Nicola Franco, Jeanette Miriam Lorenz, Stephan Günnemann
Native Robustness Quantum Machine Learning Quantum Circuit Robustness Certification Certifiable Robustness Channel Noise

June 4, 2024

Effects of Exponential Gaussian Distribution on (Double Sampling) Randomized Smoothing
Youwei Shu, Xi Xiao, Derui Wang, Yuxin Cao, Siji Chen, Jason Xue, Linyi Li, Bo Li
Mixed Effect Randomized Smoothing Gaussian Distribution Robustness Certification Smoothing Factor Norm Attack

May 14, 2024

Certifying Robustness of Graph Convolutional Networks for Node Perturbation with Polyhedra Abstract Interpretation
Boqi Chen, Kristóf Marussy, Oszkár Semeráth, Gunter Mussbacher, Dániel Varró
Adversarial Attack Native Robustness Adversarial Perturbation Robustness Certification Graph Perturbation Robustness Bound

May 13, 2024

CrossCert: A Cross-Checking Detection Approach to Patch Robustness Certification for Deep Learning Models
Qilin Zhou, Zhengyuan Wei, Haipeng Wang, Bo Jiang, W. K. Chan
Deep Learning Model Accurate Detection Adversarial Patch Attack Robustness Certification Patch Attack

March 4, 2024

COMMIT: Certifying Robustness of Multi-Sensor Fusion Systems against Semantic Attacks
Zijian Huang, Wenda Chu, Linyi Li, Chejian Xu, Bo Li
Native Robustness Multi Sensor Fusion Robustness Certification Autonomous Driving System Robust Autonomous Driving Semantic Attack CoMMIT Selection

December 18, 2023

The Pros and Cons of Adversarial Robustness
Yacine Izza, Joao Marques-Silva
Native Robustness Adversarial Example Adversarial Robustness Robustness Evaluation Robustness Certification Machine Learning Robustness

November 28, 2023

STR-Cert: Robustness Certification for Deep Text Recognition on Deep Learning Pipelines and Vision Transformers
Daqian Shao, Lukas Fesser, Marta Kwiatkowska
Convolutional Neural Network Vision Transformer Text Recognition Scene Text Recognition Adversarial Input Robustness Certification

October 13, 2023

Is Certifying $\ell_p$ Robustness Still Worthwhile?
Ravi Mangal, Klas Leino, Zifan Wang, Kai Hu, Weicheng Yu, Corina Pasareanu, Anupam Datta, Matt Fredrikson
Adversarial Attack Threat Model Robustness Certification Certified Robustness Worth Multiple Word Norm Attack \Ell_p$ Robustness

October 12, 2023

Provably Robust Cost-Sensitive Learning via Randomized Smoothing
Yuan Xin, Michael Backes, Xiao Zhang
Adversarial Perturbation Robust Learning Randomized Smoothing Robustness Certification Cost Sensitive

September 22, 2023

Pixel-wise Smoothing for Certified Robustness against Camera Motion Perturbations
Hanjiang Hu, Zuxin Liu, Linyi Li, Jiacheng Zhu, Ding Zhao
Native Robustness Robustness Certification Certified Robustness Image Smoothing

August 1, 2023

A Majority Invariant Approach to Patch Robustness Certification for Deep Learning Models
Qilin Zhou, Zhengyuan Wei, Haipeng Wang, W. K. Chan
Deep Learning Model Simple Classifier Patch Based Robustness Certification Label Set

June 27, 2023

[Re] Double Sampling Randomized Smoothing
Aryan Gupta, Sarthak Gupta, Abhay Kumar, Harsh Dugar
Native Robustness Adversarial Training Adversarial Perturbation Randomized Smoothing Robustness Certification

June 25, 2023

Computational Asymmetries in Robust Classification
Samuele Marro, Michele Lombardi
Adversarial Attack Native Robustness Adversarial Robustness Information Asymmetry Robust Classification Robustness Certification Robustness Certificate

June 19, 2023

BNN-DP: Robustness Certification of Bayesian Neural Networks via Dynamic Programming
Steven Adams, Andrea Patane, Morteza Lahijanian, Luca Laurenti
Adversarial Robustness Bayesian Neural Network Dynamic Programming Robustness Certification

June 11, 2023

Precise and Generalized Robustness Certification for Neural Networks
Yuanyuan Yuan, Shuai Wang, Zhendong Su
Neural Network Robustness Certification Neural Network Robustness Semantic Image Editing

May 31, 2023

Incremental Randomized Smoothing Certification
Shubham Ugare, Tarun Suresh, Debangshu Banerjee, Gagandeep Singh, Sasa Misailovic
Robustness Guarantee Robustness Certification Smoothed Classifier Robustness Certificate

February 4, 2023

Certified Robust Control under Adversarial Perturbations
Jinghan Yang, Hunmin Kim, Wenbin Wan, Naira Hovakimyan, Yevgeniy Vorobeychik
Native Robustness Adversarial Perturbation Robust Control Robustness Certification

January 29, 2023

Unlocking Deterministic Robustness Certification on ImageNet
Kai Hu, Andy Zou, Zifan Wang, Klas Leino, Matt Fredrikson
Supervised ImageNet Robust Learning Robust Training Robustness Certification

January 21, 2023

Limitations of Piecewise Linearity for Efficient Robustness Certification
Klas Leino
Fundamental Limitation Robust Generalization Piecewise Linear Free Counterpart Robustness Certification Smooth Activation Function

December 22, 2022

Certified Policy Smoothing for Cooperative Multi-Agent Reinforcement Learning
Ronghui Mu, Wenjie Ruan, Leandro Soriano Marcolino, Gaojie Jin, Qiang Ni
Cooperative Multi Agent Reinforcement Learning Single Agent Robustness Certification MARL Algorithm