Neural Network Verification is a Programming Language Challenge

Lucas C. Cordeiro; Matthew L.   Daggitt; Julien  Girard-Satabin; Omri  Isac; Taylor T.   Johnson; Guy  Katz; Ekaterina  Komendantskaya; Augustin   Lemesle; Edoardo Manino; Artjoms  Šinkarovs; Haoze   Wu

Neural Network Verification is a Programming Language Challenge

Lucas C. Cordeiro, Matthew L. Daggitt, Julien Girard-Satabin, Omri Isac, Taylor T. Johnson, Guy Katz, Ekaterina Komendantskaya, Augustin Lemesle, Edoardo Manino, Artjoms Šinkarovs, Haoze Wu

Systems and Software Security

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

Neural network verification is a new and rapidly developing field of research. So far, the main priority has been establishing efficient verification algorithms and tools, while proper support from the programming language perspective has been considered secondary or unimportant. Yet, there is mounting evidence that insights from the programming language community may make a difference in the future development of this domain. In this paper, we formulate neural network verification challenges as programming language challenges and suggest possible future
solutions.

Original language	English
Title of host publication	34th European Symposium on Programming
Publication status	Accepted/In press - 19 Dec 2024

Keywords

Neural Networks
Verification
Domain Specific Languages

Embargoed Document

esop2025
Accepted author manuscript, 631 KB
Licence: CC BY
Embargo ends: 17/01/26

EnnCore: End-to-End Conceptual Guarding of Neural Architectures
Cordeiro, L. (PI), Brown, G. (CoI), Freitas, A. (CoI), Luján, M. (CoI) & Mustafa, M. (CoI)
1/02/21 → 31/12/25
Project: Research

Cite this

@inproceedings{3f9d5c0e3c734b5bbf06e94a1a88e00c,

title = "Neural Network Verification is a Programming Language Challenge",

abstract = "Neural network verification is a new and rapidly developing field of research. So far, the main priority has been establishing efficient verification algorithms and tools, while proper support from the programming language perspective has been considered secondary or unimportant. Yet, there is mounting evidence that insights from the programming language community may make a difference in the future development of this domain. In this paper, we formulate neural network verification challenges as programming language challenges and suggest possible futuresolutions.",

keywords = "Neural Networks, Verification, Domain Specific Languages",

author = "Cordeiro, {Lucas C.} and Daggitt, {Matthew L.} and Julien Girard-Satabin and Omri Isac and Johnson, {Taylor T.} and Guy Katz and Ekaterina Komendantskaya and Augustin Lemesle and Edoardo Manino and Artjoms {\v S}inkarovs and Haoze Wu",

year = "2024",

month = dec,

day = "19",

language = "English",

booktitle = "34th European Symposium on Programming",

}

TY - GEN

T1 - Neural Network Verification is a Programming Language Challenge

AU - Cordeiro, Lucas C.

AU - Daggitt, Matthew L.

AU - Girard-Satabin, Julien

AU - Isac, Omri

AU - Johnson, Taylor T.

AU - Katz, Guy

AU - Komendantskaya, Ekaterina

AU - Lemesle, Augustin

AU - Manino, Edoardo

AU - Šinkarovs, Artjoms

AU - Wu, Haoze

PY - 2024/12/19

Y1 - 2024/12/19

N2 - Neural network verification is a new and rapidly developing field of research. So far, the main priority has been establishing efficient verification algorithms and tools, while proper support from the programming language perspective has been considered secondary or unimportant. Yet, there is mounting evidence that insights from the programming language community may make a difference in the future development of this domain. In this paper, we formulate neural network verification challenges as programming language challenges and suggest possible futuresolutions.

AB - Neural network verification is a new and rapidly developing field of research. So far, the main priority has been establishing efficient verification algorithms and tools, while proper support from the programming language perspective has been considered secondary or unimportant. Yet, there is mounting evidence that insights from the programming language community may make a difference in the future development of this domain. In this paper, we formulate neural network verification challenges as programming language challenges and suggest possible futuresolutions.

KW - Neural Networks

KW - Verification

KW - Domain Specific Languages

M3 - Conference contribution

BT - 34th European Symposium on Programming

ER -

Neural Network Verification is a Programming Language Challenge

Abstract

Keywords

Embargoed Document

Fingerprint

Projects

EnnCore: End-to-End Conceptual Guarding of Neural Architectures

Cite this