Exploring Trade-offs Between Black-Box and Glass-Box Models in Face Similarity: Siamese Networks vs. KNN

Abhijith Jyothi Jayachandran, Tonu James, Rajesh Jaiswal, Keith Quille

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

Abstract

This study compares black-box and glass-box models for face verification, specifically using Siamese neural networks and K-Nearest Neighbors (KNN). Using the Olivetti dataset, we showed that a Siamese network with VGG19 transfer learning reached 95.4% accuracy, outperforming a custom model's initial 49.61% accuracy. Contrastive loss improved training, while KNN known for its simplicity and interpretability achieved 91.25% accuracy but faced challenges with high-dimensional data. Grad-CAM and saliency maps provided interpretability for the Siamese network and KNN, respectively. This work underscores the trade-off between performance and explainability in these models.

Original languageEnglish
Title of host publicationHCAI-ep 2024 - Proceedings of the 2024 Conference on Human Centered Artificial Intelligence - Education and Practice
PublisherAssociation for Computing Machinery (ACM)
Pages62
Number of pages1
ISBN (Electronic)9798400711596
DOIs
Publication statusPublished - 2 Dec 2024
Event2nd Conference on Human Centered Artificial Intelligence - Education and Practice, HCAI-ep 2024 - Naples, Italy
Duration: 1 Dec 20242 Dec 2024

Publication series

NameACM International Conference Proceeding Series

Conference

Conference2nd Conference on Human Centered Artificial Intelligence - Education and Practice, HCAI-ep 2024
Country/TerritoryItaly
CityNaples
Period1/12/242/12/24

Fingerprint

Dive into the research topics of 'Exploring Trade-offs Between Black-Box and Glass-Box Models in Face Similarity: Siamese Networks vs. KNN'. Together they form a unique fingerprint.

Cite this