{"id":{"repo_id":"vt","oai_identifier":"oai:vtechworks.lib.vt.edu:10919/83714"},"canonical_url":"https://search.dev.ndltd.org/etd/vt/oai:vtechworks.lib.vt.edu:10919/83714","repository":{"repo_id":"vt","name":"Virginia Tech","base_url":"https://vtechworks.lib.vt.edu/oai/request"},"display":{"title":"Enhanced Neural Network Training Using Selective Backpropagation and Forward Propagation","abstract":"Neural networks are making headlines every day as the tool of the future, powering artificial intelligence programs and supporting technologies never seen before. However, the training of neural networks can take days or even weeks for bigger networks, and requires the use of super computers and GPUs in academia and industry in order to achieve state of the art results. This thesis discusses employing selective measures to determine when to backpropagate and forward propagate in order to reduce training time while maintaining classification performance. This thesis tests these new algorithms on the MNIST and CASIA datasets, and achieves successful results with both algorithms on the two datasets. The selective backpropagation algorithm shows a reduction of up to 93.3% of backpropagations completed, and the selective forward propagation algorithm shows a reduction of up to 72.90% in forward propagations and backpropagations completed compared to baseline runs of always forward propagating and backpropagating. This work also discusses employing the selective backpropagation algorithm on a modified dataset with disproportional under-representation of some classes compared to others.","abstract_html":"Neural networks are making headlines every day as the tool of the future, powering artificial intelligence programs and supporting technologies never seen before. However, the training of neural networks can take days or even weeks for bigger networks, and requires the use of super computers and GPUs in academia and industry in order to achieve state of the art results. This thesis discusses employing selective measures to determine when to backpropagate and forward propagate in order to reduce training time while maintaining classification performance. This thesis tests these new algorithms on the MNIST and CASIA datasets, and achieves successful results with both algorithms on the two datasets. The selective backpropagation algorithm shows a reduction of up to 93.3% of backpropagations completed, and the selective forward propagation algorithm shows a reduction of up to 72.90% in forward propagations and backpropagations completed compared to baseline runs of always forward propagating and backpropagating. This work also discusses employing the selective backpropagation algorithm on a modified dataset with disproportional under-representation of some classes compared to others.","abstract_has_math":false,"creators":["Bendelac, Shiri"],"institution":"Virginia Tech","degree_name":"Master of Science","degree_level":"masters","degree_discipline":"Computer Engineering","degree_department":"Electrical and Computer Engineering","school":null,"contributors":[],"advisors":[],"committee_chairs":["Ernst, Joseph M.","Huang, Jia-Bin"],"committee_members":["Wyatt, Chris L.","Headley, William C."],"year":2018,"date_issued":"2018-06-22","date_published":"2018-06-22","updated_at":"2026-07-22T22:19:26Z","subjects":["Machine learning","neural networks","convolutional neural networks","backpropagation","forward propagation","training"],"languages":[],"rights":["In Copyright"],"rights_urls":["http://rightsstatements.org/vocab/InC/1.0/"],"identifier_entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:15455"],"render_values":[{"text":"vt_gsexam:15455","href":null,"code":true}]}]},"links":{"outbound_url":"http://hdl.handle.net/10919/83714","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.committeechair","label":"Committee Chair","values":["Ernst, Joseph M.","Huang, Jia-Bin"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Wyatt, Chris L.","Headley, William C."]},{"key":"dc:contributor.department","label":"Department","values":["Electrical and Computer Engineering"]},{"key":"dc:creator","label":"Author","values":["Bendelac, Shiri"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2018-06-23T08:00:16Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2018-06-23T08:00:16Z"]},{"key":"dc:date.issued","label":"Date","values":["2018-06-22"]},{"key":"dc:publisher","label":"Institution","values":["Virginia Tech"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Virginia Polytechnic Institute and State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Machine learning","neural networks","convolutional neural networks","backpropagation","forward propagation","training"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://rightsstatements.org/vocab/InC/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:15455"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/10919/83714"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Neural networks are making headlines every day as the tool of the future, powering artificial intelligence programs and supporting technologies never seen before. However, the training of neural networks can take days or even weeks for bigger networks, and requires the use of super computers and GPUs in academia and industry in order to achieve state of the art results. This thesis discusses employing selective measures to determine when to backpropagate and forward propagate in order to reduce training time while maintaining classification performance. This thesis tests these new algorithms on the MNIST and CASIA datasets, and achieves successful results with both algorithms on the two datasets. The selective backpropagation algorithm shows a reduction of up to 93.3% of backpropagations completed, and the selective forward propagation algorithm shows a reduction of up to 72.90% in forward propagations and backpropagations completed compared to baseline runs of always forward propagating and backpropagating. This work also discusses employing the selective backpropagation algorithm on a modified dataset with disproportional under-representation of some classes compared to others."]},{"key":"dc:description.abstractgeneral","label":"General Abstract","values":["Neural Networks are some of the most commonly used and best performing tools in machine learning. However, training them to perform well is a tedious task that can take days or even weeks, since bigger networks perform better but take exponentially longer to train. What can be done to reduce training time? Imagine a student studying for a test. The student likely solves practice problems that cover the different topics that may be covered on the test. The student then evaluates which topics he/she knew well, and forgoes extensive practice and review on those in favor of focusing on topics he/she missed or was not as confident on. This thesis discusses following a similar approach in training neural networks in order to reduce their training time needed to achieve desired performance levels."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Master of Science"]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["ETD"]},{"key":"dc:title","label":"Title","values":["Enhanced Neural Network Training Using Selective Backpropagation and Forward Propagation"]}]}],"canonical_facts":{"dc:contributor.committeechair":["Ernst, Joseph M.","Huang, Jia-Bin"],"dc:contributor.committeemember":["Wyatt, Chris L.","Headley, William C."],"dc:contributor.department":["Electrical and Computer Engineering"],"dc:creator":["Bendelac, Shiri"],"dc:date.accessioned":["2018-06-23T08:00:16Z"],"dc:date.available":["2018-06-23T08:00:16Z"],"dc:date.issued":["2018-06-22"],"dc:description.abstract":["Neural networks are making headlines every day as the tool of the future, powering artificial intelligence programs and supporting technologies never seen before. However, the training of neural networks can take days or even weeks for bigger networks, and requires the use of super computers and GPUs in academia and industry in order to achieve state of the art results. This thesis discusses employing selective measures to determine when to backpropagate and forward propagate in order to reduce training time while maintaining classification performance. This thesis tests these new algorithms on the MNIST and CASIA datasets, and achieves successful results with both algorithms on the two datasets. The selective backpropagation algorithm shows a reduction of up to 93.3% of backpropagations completed, and the selective forward propagation algorithm shows a reduction of up to 72.90% in forward propagations and backpropagations completed compared to baseline runs of always forward propagating and backpropagating. This work also discusses employing the selective backpropagation algorithm on a modified dataset with disproportional under-representation of some classes compared to others."],"dc:description.abstractgeneral":["Neural Networks are some of the most commonly used and best performing tools in machine learning. However, training them to perform well is a tedious task that can take days or even weeks, since bigger networks perform better but take exponentially longer to train. What can be done to reduce training time? Imagine a student studying for a test. The student likely solves practice problems that cover the different topics that may be covered on the test. The student then evaluates which topics he/she knew well, and forgoes extensive practice and review on those in favor of focusing on topics he/she missed or was not as confident on. This thesis discusses following a similar approach in training neural networks in order to reduce their training time needed to achieve desired performance levels."],"dc:description.degree":["Master of Science"],"dc:format.medium":["ETD"],"dc:identifier.other":["vt_gsexam:15455"],"dc:identifier.uri":["http://hdl.handle.net/10919/83714"],"dc:publisher":["Virginia Tech"],"dc:rights":["In Copyright"],"dc:rights.uri":["http://rightsstatements.org/vocab/InC/1.0/"],"dc:subject":["Machine learning","neural networks","convolutional neural networks","backpropagation","forward propagation","training"],"dc:title":["Enhanced Neural Network Training Using Selective Backpropagation and Forward Propagation"],"dc:type":["Thesis"],"thesis:degree_discipline":["Computer Engineering"],"thesis:degree_level":["masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Virginia Polytechnic Institute and State University"]},"updated_at":"2026-07-22T22:19:26Z"}