{"id":{"repo_id":"ets-quebec","oai_identifier":"oai:espace.etsmtl.ca:3287"},"canonical_url":"https://search.ndltd.org/etd/ets-quebec/oai:espace.etsmtl.ca:3287","repository":{"repo_id":"ets-quebec","name":"ETS (Quebec)","base_url":"https://espace.etsmtl.ca/cgi/oai2"},"display":{"title":"Large-margin representation learning for small-size datasets","abstract":"This work presents a novel approach combining convolutional layers (CLs) and large-margin metric learning for training supervised models on small-size datasets. This approach has four main components: i) a set of CLs with a global average pooling; ii) an instance selection method to choose anchors and instances of interest; iii) a loss function to induce the weight update of the CLs; iv) a large-margin discriminant. Images are provided to the CLs to generate a latent representation and then used to train a large-margin discriminant. The information of the discriminant (support vectors and predictions) aids in selecting anchors and instances of interest to build a loss function that aims to minimize the distance between chosen samples. The weight update using the loss function seeks to change the latent representation to increase the margin between classes. The proposed method’s advantage is that it can train a filter bank with a small amount of data due to the reduced number of parameters. It also has reduced training costs since only a subset of instances is used in the backpropagation algorithm. Experimental results with a synthetic dataset, a texture image dataset, and three histopathologic image datasets (with textural characteristics) showed that the proposed method converges within a few epochs. It also has a low computational cost, produces close or superior accuracy results than equivalent methods, deals well with imbalanced data, and can adapt the filters to different textures.","abstract_html":"This work presents a novel approach combining convolutional layers (CLs) and large-margin metric learning for training supervised models on small-size datasets. This approach has four main components: i) a set of CLs with a global average pooling; ii) an instance selection method to choose anchors and instances of interest; iii) a loss function to induce the weight update of the CLs; iv) a large-margin discriminant. Images are provided to the CLs to generate a latent representation and then used to train a large-margin discriminant. The information of the discriminant (support vectors and predictions) aids in selecting anchors and instances of interest to build a loss function that aims to minimize the distance between chosen samples. The weight update using the loss function seeks to change the latent representation to increase the margin between classes. The proposed method’s advantage is that it can train a filter bank with a small amount of data due to the reduced number of parameters. It also has reduced training costs since only a subset of instances is used in the backpropagation algorithm. Experimental results with a synthetic dataset, a texture image dataset, and three histopathologic image datasets (with textural characteristics) showed that the proposed method converges within a few epochs. It also has a low computational cost, produces close or superior accuracy results than equivalent methods, deals well with imbalanced data, and can adapt the filters to different textures.","abstract_has_math":false,"creators":["de Matos, Jonathan"],"institution":"École de technologie supérieure","degree_name":"Thèse de doctorat électronique","degree_level":"phd","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023-07-11","date_published":"2023-07-11","updated_at":"2026-11-18T22:09:57Z","subjects":["discriminant à large-marge, couches convolutives, points d’ancrage, fonction de perte, images de texture"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["INSERT-YOUR-LIBRARY-ID-HERE3287"],"render_values":[{"text":"INSERT-YOUR-LIBRARY-ID-HERE3287","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null,"record_url":null,"record_label":null,"record_kind":null,"record_source":null,"download_url":"https://espace.etsmtl.ca/id/eprint/3287/1/DE_%20MATOS_Jonathan.pdf","download_label":"Download PDF","download_source":"dc:identifier","access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://espace.etsmtl.ca/cgi/oai2?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Aespace.etsmtl.ca%3A3287","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["de Matos, Jonathan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2023-07-11"]},{"key":"dc:publisher","label":"Institution","values":["École de technologie supérieure"]},{"key":"dc:type","label":"Dc Type","values":["Electronic Thesis or Dissertation"]},{"key":"thesis:degree_level","label":"Degree Level","values":["phd"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Thèse de doctorat électronique"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["École de technologie supérieure"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["discriminant à large-marge, couches convolutives, points d’ancrage, fonction de perte, images de texture"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["INSERT-YOUR-LIBRARY-ID-HERE3287","https://espace.etsmtl.ca/id/eprint/3287/1/DE_%20MATOS_Jonathan.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["This work presents a novel approach combining convolutional layers (CLs) and large-margin metric learning for training supervised models on small-size datasets. This approach has four main components: i) a set of CLs with a global average pooling; ii) an instance selection method to choose anchors and instances of interest; iii) a loss function to induce the weight update of the CLs; iv) a large-margin discriminant. Images are provided to the CLs to generate a latent representation and then used to train a large-margin discriminant. The information of the discriminant (support vectors and predictions) aids in selecting anchors and instances of interest to build a loss function that aims to minimize the distance between chosen samples. The weight update using the loss function seeks to change the latent representation to increase the margin between classes. The proposed method’s advantage is that it can train a filter bank with a small amount of data due to the reduced number of parameters. It also has reduced training costs since only a subset of instances is used in the backpropagation algorithm. Experimental results with a synthetic dataset, a texture image dataset, and three histopathologic image datasets (with textural characteristics) showed that the proposed method converges within a few epochs. It also has a low computational cost, produces close or superior accuracy results than equivalent methods, deals well with imbalanced data, and can adapt the filters to different textures."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Large-margin representation learning for small-size datasets"]}]}],"canonical_facts":{"dc:creator":["de Matos, Jonathan"],"dc:date":["2023-07-11"],"dc:description":["This work presents a novel approach combining convolutional layers (CLs) and large-margin metric learning for training supervised models on small-size datasets. This approach has four main components: i) a set of CLs with a global average pooling; ii) an instance selection method to choose anchors and instances of interest; iii) a loss function to induce the weight update of the CLs; iv) a large-margin discriminant. Images are provided to the CLs to generate a latent representation and then used to train a large-margin discriminant. The information of the discriminant (support vectors and predictions) aids in selecting anchors and instances of interest to build a loss function that aims to minimize the distance between chosen samples. The weight update using the loss function seeks to change the latent representation to increase the margin between classes. The proposed method’s advantage is that it can train a filter bank with a small amount of data due to the reduced number of parameters. It also has reduced training costs since only a subset of instances is used in the backpropagation algorithm. Experimental results with a synthetic dataset, a texture image dataset, and three histopathologic image datasets (with textural characteristics) showed that the proposed method converges within a few epochs. It also has a low computational cost, produces close or superior accuracy results than equivalent methods, deals well with imbalanced data, and can adapt the filters to different textures."],"dc:format":["application/pdf"],"dc:identifier":["INSERT-YOUR-LIBRARY-ID-HERE3287","https://espace.etsmtl.ca/id/eprint/3287/1/DE_%20MATOS_Jonathan.pdf"],"dc:language":["en"],"dc:publisher":["École de technologie supérieure"],"dc:subject":["discriminant à large-marge, couches convolutives, points d’ancrage, fonction de perte, images de texture"],"dc:title":["Large-margin representation learning for small-size datasets"],"dc:type":["Electronic Thesis or Dissertation"],"thesis:degree_level":["phd"],"thesis:degree_name":["Thèse de doctorat électronique"],"thesis:institution_name":["École de technologie supérieure"]},"updated_at":"2026-11-18T22:09:57Z"}