Formal Local Implication Between Two Neural Networks
(2025)- Abstract
- Given two neural network classifiers with the same in-
put and output domains, our goal is to compare the two networks
in relation to each other over an entire input region (e.g., within a
vicinity of an input sample). To this end, we establish the foundation
of formal local implication between two networks, i.e., N2
D=⇒ N1, in an entire input region D. That is, network N1 consistently makes
a correct decision every time network N2 does, and it does so in an
entire input region D. We further propose a sound formulation for
establishing such formally-verified (provably correct) local implica-
tions. The proposed formulation is relevant in the context of several
application domains, e.g., for comparing... (More) - Given two neural network classifiers with the same in-
put and output domains, our goal is to compare the two networks
in relation to each other over an entire input region (e.g., within a
vicinity of an input sample). To this end, we establish the foundation
of formal local implication between two networks, i.e., N2
D=⇒ N1, in an entire input region D. That is, network N1 consistently makes
a correct decision every time network N2 does, and it does so in an
entire input region D. We further propose a sound formulation for
establishing such formally-verified (provably correct) local implica-
tions. The proposed formulation is relevant in the context of several
application domains, e.g., for comparing a trained network and its
corresponding compact (e.g., pruned, quantized, distilled) networks.
We evaluate our formulation based on the MNIST, CIFAR10, and two
real-world medical datasets, to show its relevance (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/263d33a6-4271-46de-91af-97551a4bd864
- author
- Baninajjar, Anahita
LU
; Rezine, Ahmed
and Aminifar, Amir
LU
- organization
-
- LTH Profile Area: Engineering Health
- Secure and Networked Systems
- ELLIIT: the Linköping-Lund initiative on IT and mobile communication
- NEXTG2COM – a Vinnova Competence Centre in Advanced Digitalisation
- LTH Profile Area: Water
- LU Profile Area: Natural and Artificial Cognition
- LTH Profile Area: AI and Digitalization
- publishing date
- 2025
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- in press
- subject
- host publication
- European Conference on Artificial Intelligence (ECAI)
- pages
- 9 pages
- language
- English
- LU publication?
- yes
- id
- 263d33a6-4271-46de-91af-97551a4bd864
- date added to LUP
- 2025-09-04 23:46:24
- date last changed
- 2025-09-11 11:46:42
@inproceedings{263d33a6-4271-46de-91af-97551a4bd864,
abstract = {{Given two neural network classifiers with the same in-<br/>put and output domains, our goal is to compare the two networks<br/>in relation to each other over an entire input region (e.g., within a<br/>vicinity of an input sample). To this end, we establish the foundation<br/>of formal local implication between two networks, i.e., N2<br/>D=⇒ N1, in an entire input region D. That is, network N1 consistently makes<br/>a correct decision every time network N2 does, and it does so in an<br/>entire input region D. We further propose a sound formulation for<br/>establishing such formally-verified (provably correct) local implica-<br/>tions. The proposed formulation is relevant in the context of several<br/>application domains, e.g., for comparing a trained network and its<br/>corresponding compact (e.g., pruned, quantized, distilled) networks.<br/>We evaluate our formulation based on the MNIST, CIFAR10, and two<br/>real-world medical datasets, to show its relevance}},
author = {{Baninajjar, Anahita and Rezine, Ahmed and Aminifar, Amir}},
booktitle = {{European Conference on Artificial Intelligence (ECAI)}},
language = {{eng}},
title = {{Formal Local Implication Between Two Neural Networks}},
url = {{https://lup.lub.lu.se/search/files/227035490/FLI.pdf}},
year = {{2025}},
}