SVD-based Peephole and Clustering to Enhance Trustworthiness in DNN Classifiers

Livia Manovi*, Lorenzo Capelli, Alex Marchioni, Filippo Martinini, Gianluca Setti, Mauro Mangia, Riccardo Rovatti

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Deep Neural Networks have demonstrated impressive capabilities across various domains, yet their inherent complexity often obscures the rationale behind their predictions. This opacity poses challenges in domains where explainability is critical. Here, we present a novel methodology inspired by signal processing that leverages Singular Value Decomposition to both remove the redundancy in the neural network and derive compressed feature representations to be analyzed with clustering. We carried out empirical experiments with a network of the VGG family trained on CIFAR-10 and FMNIST datasets, and propose two strategies to address the trustworthiness issue in AI decisions.

Original languageEnglish (US)
Title of host publication2024 IEEE 6th International Conference on AI Circuits and Systems, AICAS 2024 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages129-133
Number of pages5
ISBN (Electronic)9798350383638
DOIs
StatePublished - 2024
Event6th IEEE International Conference on AI Circuits and Systems, AICAS 2024 - Abu Dhabi, United Arab Emirates
Duration: Apr 22 2024Apr 25 2024

Publication series

Name2024 IEEE 6th International Conference on AI Circuits and Systems, AICAS 2024 - Proceedings

Conference

Conference6th IEEE International Conference on AI Circuits and Systems, AICAS 2024
Country/TerritoryUnited Arab Emirates
CityAbu Dhabi
Period04/22/2404/25/24

Keywords

  • clustering
  • deep neural networks
  • interpretability
  • singular value decomposition
  • trustworthiness

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Electrical and Electronic Engineering
  • Instrumentation

Fingerprint

Dive into the research topics of 'SVD-based Peephole and Clustering to Enhance Trustworthiness in DNN Classifiers'. Together they form a unique fingerprint.

Cite this