Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/151211
Title: Contributions to explainable deep learning models
Author: Adhane, Gereziher  
Director: Masip Rodó, David  
Dehshibi, Mohammad Mahdi  
Abstract: In this work, we propose techniques to enhance the performance and transparency of convolutional neural networks (CNNs). We introduce novel methods for informative sample selection (ISS), uncertainty quantification, and visual explanation. The two ISS methods involve using reinforcement learning to filter out samples that could lead to overfitting and bias, and employing Monte Carlo dropout to estimate model uncertainty during training and inference. In addition, we present two visual explainability techniques: ADVISE, which generates detailed visual explanations and quantifies the relevance of feature map units, and UniCAM, which explains the opaque nature of knowledge distillation. These methods aim to improve model accuracy, robustness, fairness, and explainability, contributing to both academic research and the transparency of CNNs in computer vision applications.
Keywords: explainable AI
transparency
model uncertainty
sample selection
visual explainability
Document type: info:eu-repo/semantics/doctoralThesis
Issue Date: 10-Jul-2024
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Tesis doctorals

Files in This Item:
File Description SizeFormat 
PhD_Thesis.pdfGereziher_dissertation13,54 MBAdobe PDFThumbnail
View/Open
Share:
Export:
View statistics

This item is licensed under aCreative Commons License Creative Commons