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

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 sign...

Full description

Saved in:
Bibliographic Details
Main Authors: Manovi, Livia, Capelli, Lorenzo, Marchioni, Alex, Martinini, Filippo, Setti, Gianluca, Mangia, Mauro, Rovatti, Riccardo
Format: Conference Proceeding
Language:English
Subjects:
Artificial intelligence
Artificial neural networks
Circuits and systems
clustering
Complexity theory
deep neural networks
interpretability
Redundancy
Signal processing
singular value decomposition
trustworthiness
Vectors
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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.
ISSN:2834-9857
DOI:10.1109/AICAS59952.2024.10595919