Model-Agnostic Reachability Analysis on Deep Neural Networks
Not on the map. Why?
Abstract (arXiv)
Verification plays an essential role in the formal analysis of safety-critical systems. Most current verification methods have specific requirements when working on Deep Neural Networks (DNNs). They either target one particular network category, e.g., Feedforward Neural Networks (FNNs), or networks with specific activation functions, e.g., RdLU. In this paper, we develop a model-agnostic verification framework, called DeepAgn, and show that it can be applied to FNNs, Recurrent Neural Networks (RNNs), or a mixture of both. Under the assumption of Lipschitz continuity, DeepAgn analyses the reachability of DNNs based on a novel optimisation scheme with a global convergence guarantee. It does not require access to the network's internal structures, such as layers and parameters. Through reachability analysis, DeepAgn can tackle several well-known robustness problems, including computing the maximum safe radius for a given input, and generating the ground-truth adversarial examples. We also empirically demonstrate DeepAgn's superior capability and efficiency in handling a broader class of deep neural networks, including both FNNs, and RNNs with very deep layers and millions of neurons, than other state-of-the-art verification approaches.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 24 of its 36 references that are in the map. First posted on arXiv 3 Apr 2023.
Builds on
24 of 36 references.
References · 24
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2013Intriguing properties of neural networkscited by 5,114
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
- 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
- 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 739
- 2017Provable defenses against adversarial examples via the convex outer adversarial polytopecited by 564
- 2018Audio Adversarial Examples: Targeted Attacks on Speech-to-Textcited by 268
- 2016Safety Verification of Deep Neural Networkscited by 262
- 2018Towards Fast Computation of Certified Robustness for ReLU Networkscited by 254
- 2017Formal Verification of Piece-Wise Linear Feed-Forward Neural Networkscited by 241
- 2018Evaluating the Robustness of Neural Networks: An Extreme Value Theory Approachcited by 181
- 2016Crafting Adversarial Input Sequences for Recurrent Neural Networkscited by 102
12 more
- 2018Reachability Analysis of Deep Neural Networks with Provable Guaranteescited by 93
- 2017Verifying Properties of Binarized Deep Neural Networkscited by 77
- 2019POPQORN: Quantifying Robustness of Recurrent Neural Networkscited by 36
- 2018A Game-Based Approximate Verification of Deep Neural Networks with Provable Guaranteescited by 33
- 2017Output Range Analysis for Deep Neural Networkscited by 30
- 2017Crafting Adversarial Examples For Speech Paralinguistics Applicationscited by 30
- 2020Verifying Recurrent Neural Networks using Invariant Inferencecited by 25
- 2022PRoA: A Probabilistic Robustness Assessment against Functional Perturbationscited by 19
- 2022DIMBA: Discretely Masked Black-Box Attack in Single Object Trackingcited by 11
- 2023Towards Verifying the Geometric Robustness of Large-scale Neural Networkscited by 11
- 2022Certified Policy Smoothing for Cooperative Multi-Agent Reinforcement Learningcited by 7
- 20223DVerifier: Efficient Robustness Verification for 3D Point Cloud Modelscited by 5
Built on by
Citing papers per year: 2023: 1.
1 paper.