Local Robustness

Local robustness in machine learning focuses on evaluating and improving the stability of model predictions when inputs are slightly perturbed. Current research emphasizes developing verification methods and algorithms, often employing techniques like covariate perturbation analysis, mixed-integer programming, and covering designs, to certify robustness in various model architectures, including deep neural networks and binary neural networks, across different perturbation types (e.g., L0, L∞). This research is crucial for deploying machine learning models in safety-critical applications, ensuring reliable performance and mitigating the risks associated with adversarial attacks or noisy data.

Papers

August 2, 2024

Assessing Robustness of Machine Learning Models using Covariate Perturbations
Arun Prakash R, Anwesha Bhattacharyya, Joel Vaughan, Vijayan N. Nair
Adversarial Attack Native Robustness Machine Learning Model Model Robustness Noise Covariance Local Robustness

July 31, 2024

Certifying Robustness of Learning-Based Keypoint Detection and Pose Estimation Methods
Xusheng Luo, Tianhao Wei, Simin Liu, Ziwei Wang, Luis Mattei-Mendez, Taylor Loper, Joshua Neighbor, Casidhe Hutchison, Changliu Liu
Pose Estimation Keypoint Detection Keypoint Detector Certifiable Robustness Local Robustness Semantic Pose

May 31, 2024

Certifying Global Robustness for Deep Neural Networks
You Li, Guannan Zhao, Shuyu Kong, Yunqi He, Hai Zhou
Deep Neural Network Robust Deep Global Robustness Local Robustness Robustness Measurement

May 17, 2024

Boosting Few-Pixel Robustness Verification via Covering Verification Designs
Yuval Shapira, Naor Wiesel, Shahar Shabelman, Dana Drachsler-Cohen
Native Robustness Robustness Verification Local Robustness

February 29, 2024

Verification of Neural Networks' Global Robustness
Anan Kabaha, Dana Drachsler-Cohen
Neural Network Verification Task Global Robustness Local Robustness

October 31, 2023

Verification of Neural Networks Local Differential Classification Privacy
Roie Reshef, Anan Kabaha, Olga Seleznova, Dana Drachsler-Cohen
Differential Privacy Verification Task Local Differential Privacy Privacy Attack Network Structure Local Robustness

September 25, 2023

Benchmarking Local Robustness of High-Accuracy Binary Neural Networks for Enhanced Traffic Sign Recognition
Andreea Postovan, Mădălina Eraşcu
Binary Neural Network Traffic Sign Recognition Traffic Sign BNN Layer Local Robustness

May 5, 2023

Provable Preimage Under-Approximation for Neural Networks (Full Version)
Xiyue Zhang, Benjie Wang, Marta Kwiatkowska
Neural Network Neural Network Verification Robustness Analysis Local Robustness

July 13, 2022

Verifying Attention Robustness of Deep Neural Networks against Semantic Perturbations
Satoshi Munakata, Caterina Urban, Haruki Yokoyama, Koji Yamamoto, Kazuki Munakata
Deep Neural Network Saliency Map Local Robustness Accurate Saliency Semantic Perturbation Robust Attention

May 17, 2022

Hierarchical Distribution-Aware Testing of Deep Learning
Wei Huang, Xingyu Zhao, Alec Banks, Victoria Cox, Xiaowei Huang
Deep Learning Adversarial Example Adversarial Robustness Adversarial Perturbation Local Robustness Test Label Distribution Robustness Testing

May 8, 2022

VPN: Verification of Poisoning in Neural Networks
Youcheng Sun, Muhammad Usman, Divya Gopinath, Corina S. Păsăreanu
Neural Network Poisoning Attack Verification Task Data Poisoning Local Robustness Verification Tool

March 26, 2022

Efficient Global Robustness Certification of Neural Networks via Interleaving Twin-Network Encoding
Zhilu Wang, Chao Huang, Qi Zhu
Neural Network Robustness Certification Global Robustness Local Robustness Multi Encoder

March 1, 2022

A Domain-Theoretic Framework for Robustness Analysis of Neural Networks
Can Zhou, Razin A. Shaikh, Yiran Li, Amin Farjudian
Neural Network Differentiable Neural Domain Model Robustness Analysis Global Robustness Local Robustness