Reachability analysis of deep ReLU neural networks using facet-vertex incidence
Authors:
Open Access
- 19 May 2021
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Abstract
Deep Neural Networks (DNNs) are powerful machine learning models for approximating complex functions. In this work, we provide an exact reachability analysis method for DNNs with Rectified Linear Unit (ReLU) activation functions. At its core, our set-based method utilizes a facet-vertex incidence matrix, which represents a complete encoding of the combinatorial structure of convex sets. When a safety violation is detected, our approach provides backtracking which determines the complete input set that caused the safety violation. The performance of our method is evaluated and compared to other state-of-the-art methods by using the ACAS Xu flight controller and other benchmarks.Keywords
This publication has 16 references indexed in Scilit:
- Verisig2019
- Reachability analysis for neural feedback systems using regressive polynomial rule inference2019
- An abstract domain for certifying neural networksProceedings of the ACM on Programming Languages, 2019
- AI2: Safety and Robustness Certification of Neural Networks with Abstract Interpretation2018
- Output Reachable Set Estimation and Verification for Multilayer Neural NetworksIEEE Transactions on Neural Networks and Learning Systems, 2018
- Output Range Analysis for Deep Feedforward Neural Networks2018
- Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks2017
- Safety Verification of Deep Neural Networks2017
- Policy compression for aircraft collision avoidance systems2016
- An Abstraction-Refinement Approach to Verification of Artificial Neural Networks2010