{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/152728"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/152728","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"Enhancing Self-Supervised Learning through Transformations in Higher Activation Space","abstract":"We introduce Deep Augmentation, an approach to data augmentation using dropout to dynamically transform a targeted layer within a neural network, with the option to use the stop-gradient operation, offering significant improvements in model performance and generalization. We demonstrate the efficacy of Deep Augmentation through extensive experiments on contrastive learning tasks in computer vision and NLP domains, where we observe substantial performance gains with ResNets and Transformers as the underlying models. Our experimentation reveals that targeting deeper layers with Deep Augmentation outperforms augmenting the input data, and the simple network- and data-agnostic nature of this approach enables its seamless integration into computer vision and NLP pipelines.","abstract_html":"We introduce Deep Augmentation, an approach to data augmentation using dropout to dynamically transform a targeted layer within a neural network, with the option to use the stop-gradient operation, offering significant improvements in model performance and generalization. We demonstrate the efficacy of Deep Augmentation through extensive experiments on contrastive learning tasks in computer vision and NLP domains, where we observe substantial performance gains with ResNets and Transformers as the underlying models. Our experimentation reveals that targeting deeper layers with Deep Augmentation outperforms augmenting the input data, and the simple network- and data-agnostic nature of this approach enables its seamless integration into computer vision and NLP pipelines.","abstract_has_math":false,"creators":["Gabrielsson, Rickard Brüel"],"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":["Solomon, Justin"],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023-09","date_published":"2023-09","updated_at":"2026-07-22T22:21:11Z","subjects":[],"languages":[],"rights":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"],"rights_urls":["https://rightsstatements.org/page/InC-EDU/1.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/152728","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Solomon, Justin"]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Gabrielsson, Rickard Brüel"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2023-11-02T20:11:33Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2023-11-02T20:11:33Z"]},{"key":"dc:date.issued","label":"Date","values":["2023-09"]},{"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","Master of Science in Electrical Engineering and Computer Science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://rightsstatements.org/page/InC-EDU/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/152728"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["We introduce Deep Augmentation, an approach to data augmentation using dropout to dynamically transform a targeted layer within a neural network, with the option to use the stop-gradient operation, offering significant improvements in model performance and generalization. We demonstrate the efficacy of Deep Augmentation through extensive experiments on contrastive learning tasks in computer vision and NLP domains, where we observe substantial performance gains with ResNets and Transformers as the underlying models. Our experimentation reveals that targeting deeper layers with Deep Augmentation outperforms augmenting the input data, and the simple network- and data-agnostic nature of this approach enables its seamless integration into computer vision and NLP pipelines."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["S.M."]},{"key":"dc:title","label":"Title","values":["Enhancing Self-Supervised Learning through Transformations in Higher Activation Space"]}]}],"canonical_facts":{"dc:contributor.advisor":["Solomon, Justin"],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Gabrielsson, Rickard Brüel"],"dc:date.accessioned":["2023-11-02T20:11:33Z"],"dc:date.available":["2023-11-02T20:11:33Z"],"dc:date.issued":["2023-09"],"dc:description.abstract":["We introduce Deep Augmentation, an approach to data augmentation using dropout to dynamically transform a targeted layer within a neural network, with the option to use the stop-gradient operation, offering significant improvements in model performance and generalization. We demonstrate the efficacy of Deep Augmentation through extensive experiments on contrastive learning tasks in computer vision and NLP domains, where we observe substantial performance gains with ResNets and Transformers as the underlying models. Our experimentation reveals that targeting deeper layers with Deep Augmentation outperforms augmenting the input data, and the simple network- and data-agnostic nature of this approach enables its seamless integration into computer vision and NLP pipelines."],"dc:description.degree":["S.M."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/152728"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"],"dc:rights.uri":["https://rightsstatements.org/page/InC-EDU/1.0/"],"dc:title":["Enhancing Self-Supervised Learning through Transformations in Higher Activation Space"],"dc:type":["Thesis"],"thesis:degree_name":["Master","Master of Science in Electrical Engineering and Computer Science"]},"updated_at":"2026-07-22T22:21:11Z"}