Algorithms for Verifying Deep Neural Networks
Foundations and Trends® in Optimization · 2021-02-11 · 4 (3-4) · 244-404
DOI: 10.1561/2400000035 ↗At a glance
Journal article in Foundations and Trends® in Optimization, published online on 2021-02-11. Written by Changliu Liu and 5 co-authors. Cited 192 times so far (Crossref); 77 references.
Abstract (original language)
Deep neural networks are widely used for nonlinear function approximation, with applications ranging from computer vision to control. Although these networks involve the composition of simple arithmetic operations, it can be very challenging to verify whether a particular network satisfies certain input-output properties. This article surveys methods that have emerged recently for soundly verifying such properties. These methods borrow insights from reachability analysis, optimization, and search. We discuss fundamental differences and connections between existing algorithms. In addition, we provide pedagogical implementations of existing methods and compare them on a set of benchmark problems.
Cite
APA
Liu, C., Arnon, T., Lazarus, C., Strong, C., Barrett, C., & Kochenderfer, M. J. (2021). Algorithms for Verifying Deep Neural Networks. Foundations and Trends® in Optimization, 4(3-4), 244-404. https://doi.org/10.1561/2400000035
Vancouver
Liu C, Arnon T, Lazarus C, Strong C, Barrett C, Kochenderfer MJ. Algorithms for Verifying Deep Neural Networks. Foundations and Trends® in Optimization. 2021;4(3-4):244-404. doi:10.1561/2400000035
BibTeX
@article{Liu2021,
author = {Liu, Changliu and Arnon, Tomer and Lazarus, Chris and Strong, Christopher and Barrett, Clark and Kochenderfer, Mykel J.},
title = {Algorithms for Verifying Deep Neural Networks},
journal = {Foundations and Trends® in Optimization},
year = {2021},
volume = {4},
number = {3-4},
pages = {244--404},
doi = {10.1561/2400000035}
}
EUVIMED