IMPROVING UNCERTAINTY QUANTIFICATION OF DEEP CLASSIFIERS: A NEIGHBORHOOD CONFORMAL PREDICTION APPROACH

Subhankar Ghosh

doi:10.7273/000007043

Back

IMPROVING UNCERTAINTY QUANTIFICATION OF DEEP CLASSIFIERS: A NEIGHBORHOOD CONFORMAL PREDICTION APPROACH

Thesis

Open access

IMPROVING UNCERTAINTY QUANTIFICATION OF DEEP CLASSIFIERS: A NEIGHBORHOOD CONFORMAL PREDICTION APPROACH

Subhankar Ghosh

Washington State University

Master of Science (MS), Washington State University

07/2024

DOI:

https://doi.org/10.7273/000007043

Files and links (1)

pdf

Subhankar_MS_thesis (4)920.15 kBDownload View

Open Access

Abstract

Classification

Conformal Prediction

Deep Learning

Uncertainty Quantification

Safe deployment of deep neural networks in high-stake real-world applications requires theoretically sound uncertainty quantification. Conformal prediction (CP) is a principled framework for uncertainty quantification of deep models in the form of prediction set for classification tasks with a user-specified coverage (i.e., true class label is contained with high probability). This dissertation proposes a novel algorithm referred to as Neighborhood Conformal Prediction (NCP) to improve the efficiency of uncertainty quantification from CP for deep classifiers (i.e., reduce prediction set size). The key idea behind NCP is to use the learned representation of the neural network to identify k nearest-neighbors calibration examples for a given testing input and assign them importance weights proportional to their distance to create adaptive prediction sets. We theoretically show that if the learned data representation of the neural network satisfies some mild conditions, NCP will produce smaller prediction sets than traditional CP algorithms. Our comprehensive experiments on CIFAR-10, CIFAR-100, and ImageNet datasets using diverse deep neural networks strongly demonstrate that NCP leads to significant reduction in prediction set size over prior CP methods.

Metrics

7 File views/ downloads

20 Record Views

Details

Title: IMPROVING UNCERTAINTY QUANTIFICATION OF DEEP CLASSIFIERS
Creators: Subhankar Ghosh
Contributors: Yan Yan (Co-Chair)
Janardhan Rao Doppa (Co-Chair)
Diane J. Cook (Committee Member)
Ganapati Bhat (Committee Member)
Awarding Institution: Washington State University
Academic Unit: School of Electrical Engineering and Computer Science
Theses and Dissertations: Master of Science (MS), Washington State University
Publisher: Washington State University
Number of pages: 63
Identifiers: 99901152338701842
Language: English
Resource Type: Thesis