“Algorithms for Verifying Deep Neural Networks” by Changliu Liu, Tomer Arnon, Christopher Lazarus, Christopher Strong, Clark Barrett, and Mykel J. Kochenderfer. Foundations and Trends in Optimization, vol. 4, no. 3-4, Feb. 2021, pp. 244-404, now publishers.
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.
BibTeX entry:
@article{LAL+21,
author = {Changliu Liu and Tomer Arnon and Christopher Lazarus and
Christopher Strong and Clark Barrett and Mykel J. Kochenderfer},
title = {Algorithms for Verifying Deep Neural Networks},
journal = {Foundations and Trends in Optimization},
volume = {4},
number = {3-4},
pages = {244--404},
publisher = {now publishers},
month = feb,
year = {2021},
issn = {2167-3888},
doi = {10.1561/2400000035},
url = {http://theory.stanford.edu/~barrett/pubs/LAL+21.pdf}
}
(This webpage was created with bibtex2web.)