<?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>2026-11-21T07:45:52Z</responseDate><request verb="GetRecord" identifier="oai:macsphere.mcmaster.ca:11375/26127" metadataPrefix="dim">https://prod-ms-be.lib.mcmaster.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:macsphere.mcmaster.ca:11375/26127</identifier><datestamp>2026-09-24T01:38:19Z</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">Chiang, Fei</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="advisor">Zheng, Rong</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="author">Manzuri Shalmani, Shervin</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="department" lang="en_US">Computing and Software</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2021-01-03T01:29:08Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2021-01-03T01:29:08Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2020</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">http://hdl.handle.net/11375/26127</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">https://doi.org/10.71548/19553</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en_US">Modern neural networks are powerful predictive models. However, when it comes&#xd;
to recognizing that they may be wrong about their predictions and measuring the&#xd;
certainty of beliefs, they perform poorly. For one of the most common activation&#xd;
functions, the ReLU and its variants, even a well-calibrated model can produce incorrect&#xd;
but high confidence predictions. In the related task of action recognition, most&#xd;
current classification methods are based on clip-level classifiers that densely sample a&#xd;
given video for non-overlapping, same sized clips and aggregate the results using an&#xd;
aggregation function - typically averaging - to achieve video level predictions. While&#xd;
this approach has shown to be effective, it is sub-optimal in recognition accuracy&#xd;
and has a high computational overhead. To mitigate both these issues, we propose&#xd;
the confidence distillation framework to  firstly teach a representation of uncertainty&#xd;
of the teacher to the student and secondly divide the task of full video prediction&#xd;
between the student and the teacher models. We conduct extensive experiments&#xd;
on three action recognition datasets and demonstrate that our framework achieves&#xd;
state-of-the-art results in action recognition accuracy and computational efficiency.</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degreetype" lang="en_US">Thesis</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="degree" lang="en_US">Master of Science (MSc)</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="layabstract" lang="en_US">We devise a distillation loss function to train an efficient sampler/classifier for video-based action recognition tasks.</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso" lang="en_US">en</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Deep Learning</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Computer Vision</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Artificial Intelligence</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Efficient Inference</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Regularization</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Loss Function</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Machine Learning</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en_US">Distillation</dim:field>
   <dim:field mdschema="dc" element="title" lang="en_US">Confidence Distillation for Efficient Action Recognition</dim:field>
   <dim:field mdschema="dc" element="type" lang="en_US">Thesis</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
</dim:dim></metadata></record></GetRecord></OAI-PMH>