paper-with-me

Papers

Certifying Global Robustness for Deep Neural Networks

2024-05-31 · You Li, Guannan Zhao, Shuyu Kong, Yunqi He, Hai Zhou

A globally robust deep neural network resists perturbations on all meaningful inputs. Current robustness certification methods emphasize local robustness, struggling to scale and generalize. This paper presents a systematic and efficient method to evaluate and verify global robustness for deep neural networks, leveraging the PAC verification framework for solid guarantees on verification results. We utilize probabilistic programs to characterize meaningful input regions, setting a realistic standard for global robustness. Additionally, we introduce the cumulative robustness curve as a criterion in evaluating global robustness. We design a statistical method that combines multi-level splitting and regression analysis for the estimation, significantly reducing the execution time. Experimental results demonstrate the efficiency and effectiveness of our verification method and its capability to find rare and diversified counterexamples for adversarial training.

📄 PDF Abstract BibTeX arXiv:2405.20556

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothing

2021-06-17 · ICLR 2022 4 · Fan Wu, Linyi Li, Zijian Huang, Yevgeniy Vorobeychik 외

As reinforcement learning (RL) has achieved great success and been even adopted in safety-critical domains such as autonomous vehicles, a range of empirical studies have been conducted to improve its robustness against a…

Atari GamesAutonomous Vehiclesreinforcement-learningReinforcement Learning+1

Incremental Randomized Smoothing Certification

2023-05-31 · Shubham Ugare, Tarun Suresh, Debangshu Banerjee, Gagandeep Singh 외

Randomized smoothing-based certification is an effective approach for obtaining robustness certificates of deep neural networks (DNNs) against adversarial attacks. This method constructs a smoothed DNN model and certifie…

Efficient Global Robustness Certification of Neural Networks via Interleaving Twin-Network Encoding

2022-03-26 · Zhilu Wang, Chao Huang, Qi Zhu

The robustness of deep neural networks has received significant interest recently, especially when being deployed in safety-critical systems, as it is important to analyze how sensitive the model output is under input pe…

Certifying Distributional Robustness using Lipschitz Regularisation

2019-09-25 · Zac Cranko, Zhan Shi, Xinhua Zhang, Simon Kornblith 외

Distributional robust risk (DRR) minimisation has arisen as a flexible and effective framework for machine learning. Approximate solutions based on dualisation have become particularly favorable in addressing the semi-in…

Globally-Robust Neural Networks

2021-02-16 · Klas Leino, Zifan Wang, Matt Fredrikson

The threat of adversarial examples has motivated work on training certifiably robust neural networks to facilitate efficient verification of local robustness at inference time. We formalize a notion of global robustness,…