{"id":{"repo_id":"iupui","oai_identifier":"oai:scholarworks.indianapolis.iu.edu:1805/29173"},"canonical_url":"https://search.dev.ndltd.org/etd/iupui/oai:scholarworks.indianapolis.iu.edu:1805/29173","repository":{"repo_id":"iupui","name":"IUPUI","base_url":"https://scholarworks.indianapolis.iu.edu/server/oai/request"},"display":{"title":"Efficientnext: Efficientnet For Embedded Systems","abstract":"Convolutional Neural Networks have come a long way since AlexNet. Each year the limits of the state of the art are being pushed to new levels. EfficientNet pushed the performance metrics to a new high and EfficientNetV2 even more so. Even so, architectures for mobile applications can benefit from improved accuracy and reduced model footprint. The classic Inverted Residual block has been the foundation upon which most mobile networks seek to improve. EfficientNet architecture is built using the same Inverted Residual block. In this thesis we experiment with Harmonious Bottlenecks in place of the Inverted Residuals to observe a reduction in the number of parameters and improvement in accuracy. The designed network is then deployed on the NXP i.MX 8M Mini board for Image classification.","abstract_html":"Convolutional Neural Networks have come a long way since AlexNet. Each year the limits of the state of the art are being pushed to new levels. EfficientNet pushed the performance metrics to a new high and EfficientNetV2 even more so. Even so, architectures for mobile applications can benefit from improved accuracy and reduced model footprint. The classic Inverted Residual block has been the foundation upon which most mobile networks seek to improve. EfficientNet architecture is built using the same Inverted Residual block. In this thesis we experiment with Harmonious Bottlenecks in place of the Inverted Residuals to observe a reduction in the number of parameters and improvement in accuracy. The designed network is then deployed on the NXP i.MX 8M Mini board for Image classification.","abstract_has_math":false,"creators":["Deokar, Abhishek"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["El-Sharkawy, Mohamed"],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-05","date_published":"2022-05","updated_at":"2026-07-24T02:40:24Z","subjects":["Convolutional Neural Networks","Coordinate Attention","Adaptive Sharpness Aware Minimization","Harmonious Bottleneck","Sandglass Bottleneck","Inverted Residual","Fused-MBConv","CIFAR-10","NXP i.MX 8M Mini","GELU","Swish","ReLU6"],"languages":["en_US"],"rights":["Attribution 4.0 International"],"rights_urls":["http://creativecommons.org/licenses/by/4.0/"],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://dx.doi.org/10.7912/C2/2923"],"render_values":[{"text":"http://dx.doi.org/10.7912/C2/2923","href":"http://dx.doi.org/10.7912/C2/2923","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/1805/29173","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["El-Sharkawy, Mohamed"]},{"key":"dc:contributor.other","label":"Dc Contributor Other","values":["King, Brian","Rizkalla, Maher"]},{"key":"dc:creator","label":"Author","values":["Deokar, Abhishek"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2022-05-27T14:06:52Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2022-05-27T14:06:52Z"]},{"key":"dc:date.issued","label":"Date","values":["2022-05"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Convolutional Neural Networks","Coordinate Attention","Adaptive Sharpness Aware Minimization","Harmonious Bottleneck","Sandglass Bottleneck","Inverted Residual","Fused-MBConv","CIFAR-10","NXP i.MX 8M Mini","GELU","Swish","ReLU6"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://creativecommons.org/licenses/by/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1805/29173","http://dx.doi.org/10.7912/C2/2923"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Indiana University-Purdue University Indianapolis (IUPUI)"]},{"key":"dc:description.abstract","label":"Abstract","values":["Convolutional Neural Networks have come a long way since AlexNet. Each year the limits of the state of the art are being pushed to new levels. EfficientNet pushed the performance metrics to a new high and EfficientNetV2 even more so. Even so, architectures for mobile applications can benefit from improved accuracy and reduced model footprint. The classic Inverted Residual block has been the foundation upon which most mobile networks seek to improve. EfficientNet architecture is built using the same Inverted Residual block. In this thesis we experiment with Harmonious Bottlenecks in place of the Inverted Residuals to observe a reduction in the number of parameters and improvement in accuracy. The designed network is then deployed on the NXP i.MX 8M Mini board for Image classification."]},{"key":"dc:title","label":"Title","values":["Efficientnext: Efficientnet For Embedded Systems"]}]}],"canonical_facts":{"dc:contributor.advisor":["El-Sharkawy, Mohamed"],"dc:contributor.other":["King, Brian","Rizkalla, Maher"],"dc:creator":["Deokar, Abhishek"],"dc:date.accessioned":["2022-05-27T14:06:52Z"],"dc:date.available":["2022-05-27T14:06:52Z"],"dc:date.issued":["2022-05"],"dc:description":["Indiana University-Purdue University Indianapolis (IUPUI)"],"dc:description.abstract":["Convolutional Neural Networks have come a long way since AlexNet. Each year the limits of the state of the art are being pushed to new levels. EfficientNet pushed the performance metrics to a new high and EfficientNetV2 even more so. Even so, architectures for mobile applications can benefit from improved accuracy and reduced model footprint. The classic Inverted Residual block has been the foundation upon which most mobile networks seek to improve. EfficientNet architecture is built using the same Inverted Residual block. In this thesis we experiment with Harmonious Bottlenecks in place of the Inverted Residuals to observe a reduction in the number of parameters and improvement in accuracy. The designed network is then deployed on the NXP i.MX 8M Mini board for Image classification."],"dc:identifier.uri":["https://hdl.handle.net/1805/29173","http://dx.doi.org/10.7912/C2/2923"],"dc:language.iso":["en_US"],"dc:rights":["Attribution 4.0 International"],"dc:rights.uri":["http://creativecommons.org/licenses/by/4.0/"],"dc:subject":["Convolutional Neural Networks","Coordinate Attention","Adaptive Sharpness Aware Minimization","Harmonious Bottleneck","Sandglass Bottleneck","Inverted Residual","Fused-MBConv","CIFAR-10","NXP i.MX 8M Mini","GELU","Swish","ReLU6"],"dc:title":["Efficientnext: Efficientnet For Embedded Systems"],"dc:type":["Thesis"]},"updated_at":"2026-07-24T02:40:24Z"}