BERN-NN: Tight Bound Propagation For Neural Networks Using Bernstein Polynomial Interval Arithmetic

DOI10.1145/3575870.3587126arXiv2211.14438OpenAlexW4375852616MaRDI QIDQ6202105

Unnamed Author, Unnamed Author, Yasser Shoukry, Haitham Khedr

Publication date: 21 February 2024

Published in: Proceedings of the 26th ACM International Conference on Hybrid Systems: Computation and Control (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2211.14438

zbMATH Keywords

neural networksBernstein polynomialsabstraction refinement

Mathematics Subject Classification ID

Formal languages and automata (68Q45) Specification and verification (program logics, model checking, etc.) (68Q60) Control/observation systems governed by functional relations other than differential equations (such as hybrid and switching systems) (93C30)

Cites Work

• Unnamed Item
• Unnamed Item
• The Bernstein polynomial basis: a centennial retrospective
• Algebraic manipulation in the Bernstein form made simple via convolutions
• Fast construction of constant bound functions for sparse polynomials
• Algorithms for polynomials in Bernstein form
• Deep neural networks and mixed integer linear optimization
• Verification of piecewise deep neural networks: a star set approach with zonotope pre-filter
• Reluplex: an efficient SMT solver for verifying deep neural networks
• The Stone-Weierstrass Theorem
• Formal verification of neural network controlled autonomous systems
• Reachability analysis for neural feedback systems using regressive polynomial rule inference
• Verisig
• Sherlock - A tool for verification of neural network feedback systems
• Strong mixed-integer programming formulations for trained neural networks

This page was built for publication: BERN-NN: Tight Bound Propagation For Neural Networks Using Bernstein Polynomial Interval Arithmetic