{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/123067"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/123067","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"Towards understanding residual neural networks","abstract":"Residual networks (ResNets) are now a prominent architecture in the field of deep learning. However, an explanation for their success remains elusive. The original view is that residual connections allows for the training of deeper networks, but it is not clear that added layers are always useful, or even how they are used. In this work, we find that residual connections distribute learning behavior across layers, allowing resnets to indeed effectively use deeper layers and outperform standard networks. We support this explanation with results for network gradients and representation learning that show that residual connections make the training of individual residual blocks easier.","abstract_html":"Residual networks (ResNets) are now a prominent architecture in the field of deep learning. However, an explanation for their success remains elusive. The original view is that residual connections allows for the training of deeper networks, but it is not clear that added layers are always useful, or even how they are used. In this work, we find that residual connections distribute learning behavior across layers, allowing resnets to indeed effectively use deeper layers and outperform standard networks. We support this explanation with results for network gradients and representation learning that show that residual connections make the training of individual residual blocks easier.","abstract_has_math":false,"creators":["Zeng, Brandon."],"institution":"Massachusetts Institute of Technology","degree_name":"Master","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Aleksander Ma̧dry."],"committee_chairs":[],"committee_members":[],"year":2019,"date_issued":"2019","date_published":"2019","updated_at":"2026-07-22T22:21:51Z","subjects":["Electrical Engineering and Computer Science."],"languages":["eng"],"rights":["MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission."],"rights_urls":["http://dspace.mit.edu/handle/1721.1/7582"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/123067","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Aleksander Ma̧dry."]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","EECS"]},{"key":"dc:contributor.other","label":"Dc Contributor Other","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science."]},{"key":"dc:creator","label":"Author","values":["Zeng, Brandon."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2019-11-22T00:09:30Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2019-11-22T00:09:30Z"]},{"key":"dc:date.issued","label":"Date","values":["2019"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Electrical Engineering and Computer Science."]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission."]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://dspace.mit.edu/handle/1721.1/7582"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/123067"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2019","Cataloged from PDF version of thesis.","Includes bibliographical references (page 37)."]},{"key":"dc:description.abstract","label":"Abstract","values":["Residual networks (ResNets) are now a prominent architecture in the field of deep learning. However, an explanation for their success remains elusive. The original view is that residual connections allows for the training of deeper networks, but it is not clear that added layers are always useful, or even how they are used. In this work, we find that residual connections distribute learning behavior across layers, allowing resnets to indeed effectively use deeper layers and outperform standard networks. We support this explanation with results for network gradients and representation learning that show that residual connections make the training of individual residual blocks easier."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["M. Eng."]},{"key":"dc:title","label":"Title","values":["Towards understanding residual neural networks"]}]}],"canonical_facts":{"dc:contributor.advisor":["Aleksander Ma̧dry."],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","EECS"],"dc:contributor.other":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science."],"dc:creator":["Zeng, Brandon."],"dc:date.accessioned":["2019-11-22T00:09:30Z"],"dc:date.available":["2019-11-22T00:09:30Z"],"dc:date.issued":["2019"],"dc:description":["Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2019","Cataloged from PDF version of thesis.","Includes bibliographical references (page 37)."],"dc:description.abstract":["Residual networks (ResNets) are now a prominent architecture in the field of deep learning. However, an explanation for their success remains elusive. The original view is that residual connections allows for the training of deeper networks, but it is not clear that added layers are always useful, or even how they are used. In this work, we find that residual connections distribute learning behavior across layers, allowing resnets to indeed effectively use deeper layers and outperform standard networks. We support this explanation with results for network gradients and representation learning that show that residual connections make the training of individual residual blocks easier."],"dc:description.degree":["M. Eng."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/123067"],"dc:language.iso":["eng"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission."],"dc:rights.uri":["http://dspace.mit.edu/handle/1721.1/7582"],"dc:subject":["Electrical Engineering and Computer Science."],"dc:title":["Towards understanding residual neural networks"],"dc:type":["Thesis"],"thesis:degree_name":["Master"]},"updated_at":"2026-07-22T22:21:51Z"}