<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-21T03:48:08Z</responseDate><request verb="GetRecord" identifier="oai:macsphere.mcmaster.ca:11375/33107" metadataPrefix="dim">https://prod-ms-be.lib.mcmaster.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:macsphere.mcmaster.ca:11375/33107</identifier><datestamp>2026-09-24T04:05:25Z</datestamp><setSpec>com_11375_271</setSpec><setSpec>col_11375_272</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:doc="http://www.lyncode.com/xoai" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="advisor">He, Wenbo</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="author">Li, Chengqi</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="department">Computing and Software</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2026-05-07T14:57:08Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2026</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://hdl.handle.net/11375/33107</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://doi.org/10.71548/18185</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract">Deep learning systems are widely used in everyday technologies such as facial recognition, image analysis, and 3D scene reconstruction. However, these systems often face an important challenge: noise. Noise can come from imperfect data, human labeling errors, or low-quality sensors, and it can reduce the reliability of learning systems. At the same time, noise is deliberately added to protect people's privacy, especially when sensitive data such as facial images are involved. This thesis studies noise as a double-edged sword in trustworthy artificial intelligence.

The thesis proposes new methods that carefully add minimal noise to protect privacy while preserving usefulness, as well as techniques that make learning systems more robust to unavoidable noise. These methods improve face anonymization, learning from noisy data, and 3D scene reconstruction in real-world environments. Overall, this work shows how noise can be both safely used and effectively managed to build more reliable and privacy-preserving deep learning systems.</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degreetype">Thesis</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degree">Doctor of Philosophy (PhD)</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso">en</dim:field>
   <dim:field mdschema="dc" element="rights" lang="en">Attribution-NonCommercial-ShareAlike 4.0 International</dim:field>
   <dim:field mdschema="dc" element="rights" qualifier="uri">http://creativecommons.org/licenses/by-nc-sa/4.0/</dim:field>
   <dim:field mdschema="dc" element="subject">Privacy</dim:field>
   <dim:field mdschema="dc" element="subject">Robust learning</dim:field>
   <dim:field mdschema="dc" element="subject">3D vision</dim:field>
   <dim:field mdschema="dc" element="title">A Double-Edged Sword: The Role of Noise in Trustworthy Deep Learning</dim:field>
   <dim:field mdschema="dc" element="type" lang="en">Thesis</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
</dim:dim></metadata></record></GetRecord></OAI-PMH>