{"id":{"repo_id":"clemson","oai_identifier":"oai:open.clemson.edu:all_theses-3918"},"canonical_url":"https://search.ndltd.org/etd/clemson/oai:open.clemson.edu:all_theses-3918","repository":{"repo_id":"clemson","name":"Clemson University","base_url":"https://open.clemson.edu/do/oai/"},"display":{"title":"Understanding 1D Convolutional Neural Networks Using Multiclass Time-Varying Signals","abstract":"In recent times, we have seen a surge in usage of Convolutional Neural Networks to solve all kinds of problems - from handwriting recognition to object recognition and from natural language processing to detecting exoplanets. Though the technology has been around for quite some time, there is still a lot of scope to do research on what’s really happening ’under the hood’ in a CNN model. CNNs are considered to be black boxes which learn something from complex data and provides desired results. In this thesis, an eﬀort has been made to explain what exactly CNNs are learning by training the network with carefully selected input data. The data considered here are one dimensional time varying signals and hence the 1-D convolutional neural networks are used to train, test and to analyze the learned weights. The ﬁeld of digital signal processing (DSP) gives a lot of insight into understanding the seemingly random weights learned by CNN. In particular, the concepts of Fourier transform, Savitzky-Golay ﬁlters, Guassian ﬁlters and FIR ﬁlter design lights up seeming dark alley of CNNs. As a result of this study, a few interesting inferences can be made regarding dropout regularization, optimal kernel length and optimal number of convolution layers.","abstract_html":"In recent times, we have seen a surge in usage of Convolutional Neural Networks to solve all kinds of problems - from handwriting recognition to object recognition and from natural language processing to detecting exoplanets. Though the technology has been around for quite some time, there is still a lot of scope to do research on what’s really happening ’under the hood’ in a CNN model. CNNs are considered to be black boxes which learn something from complex data and provides desired results. In this thesis, an eﬀort has been made to explain what exactly CNNs are learning by training the network with carefully selected input data. The data considered here are one dimensional time varying signals and hence the 1-D convolutional neural networks are used to train, test and to analyze the learned weights. The ﬁeld of digital signal processing (DSP) gives a lot of insight into understanding the seemingly random weights learned by CNN. In particular, the concepts of Fourier transform, Savitzky-Golay ﬁlters, Guassian ﬁlters and FIR ﬁlter design lights up seeming dark alley of CNNs. As a result of this study, a few interesting inferences can be made regarding dropout regularization, optimal kernel length and optimal number of convolution layers.","abstract_has_math":false,"creators":["Srinivasamurthy, Ravisutha Sakrepatna"],"institution":null,"degree_name":"Master of Science (MS)","degree_level":"Thesis","degree_discipline":"Electrical and Computer Engineering (Holcomb Dept. of)","degree_department":null,"school":null,"contributors":["Dr. Robert J. Schalkoff, Committee Chair","Dr. Harlan B. Russell","Dr. Ilya Safro"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T07:18:20Z","subjects":["CNN","CNN and DSP","Convolutional Neural Network","Deep Learning","Machine Learning","Signal Processing"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://open.clemson.edu/all_theses/2911","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://open.clemson.edu/all_theses/2911","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.clemson.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Aopen.clemson.edu%3Aall_theses-3918","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dr. Robert J. Schalkoff, Committee Chair","Dr. Harlan B. Russell","Dr. Ilya Safro"]},{"key":"dc:creator","label":"Author","values":["Srinivasamurthy, Ravisutha Sakrepatna"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2018-09-19T07:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2018-08-01T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical and Computer Engineering (Holcomb Dept. of)"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["CNN","CNN and DSP","Convolutional Neural Network","Deep Learning","Machine Learning","Signal Processing"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://open.clemson.edu/all_theses/2911"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["In recent times, we have seen a surge in usage of Convolutional Neural Networks to solve all kinds of problems - from handwriting recognition to object recognition and from natural language processing to detecting exoplanets. Though the technology has been around for quite some time, there is still a lot of scope to do research on what’s really happening ’under the hood’ in a CNN model. CNNs are considered to be black boxes which learn something from complex data and provides desired results. In this thesis, an eﬀort has been made to explain what exactly CNNs are learning by training the network with carefully selected input data. The data considered here are one dimensional time varying signals and hence the 1-D convolutional neural networks are used to train, test and to analyze the learned weights. The ﬁeld of digital signal processing (DSP) gives a lot of insight into understanding the seemingly random weights learned by CNN. In particular, the concepts of Fourier transform, Savitzky-Golay ﬁlters, Guassian ﬁlters and FIR ﬁlter design lights up seeming dark alley of CNNs. As a result of this study, a few interesting inferences can be made regarding dropout regularization, optimal kernel length and optimal number of convolution layers."]},{"key":"dc:title","label":"Title","values":["Understanding 1D Convolutional Neural Networks Using Multiclass Time-Varying Signals"]}]}],"canonical_facts":{"dc:contributor":["Dr. Robert J. Schalkoff, Committee Chair","Dr. Harlan B. Russell","Dr. Ilya Safro"],"dc:creator":["Srinivasamurthy, Ravisutha Sakrepatna"],"dc:date.available":["2018-09-19T07:00:00Z"],"dc:date.created":["2018-08-01T07:00:00Z"],"dc:description.abstract":["In recent times, we have seen a surge in usage of Convolutional Neural Networks to solve all kinds of problems - from handwriting recognition to object recognition and from natural language processing to detecting exoplanets. Though the technology has been around for quite some time, there is still a lot of scope to do research on what’s really happening ’under the hood’ in a CNN model. CNNs are considered to be black boxes which learn something from complex data and provides desired results. In this thesis, an eﬀort has been made to explain what exactly CNNs are learning by training the network with carefully selected input data. The data considered here are one dimensional time varying signals and hence the 1-D convolutional neural networks are used to train, test and to analyze the learned weights. The ﬁeld of digital signal processing (DSP) gives a lot of insight into understanding the seemingly random weights learned by CNN. In particular, the concepts of Fourier transform, Savitzky-Golay ﬁlters, Guassian ﬁlters and FIR ﬁlter design lights up seeming dark alley of CNNs. As a result of this study, a few interesting inferences can be made regarding dropout regularization, optimal kernel length and optimal number of convolution layers."],"dc:identifier":["https://open.clemson.edu/all_theses/2911"],"dc:subject":["CNN","CNN and DSP","Convolutional Neural Network","Deep Learning","Machine Learning","Signal Processing"],"dc:title":["Understanding 1D Convolutional Neural Networks Using Multiclass Time-Varying Signals"],"thesis:degree_discipline":["Electrical and Computer Engineering (Holcomb Dept. of)"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Master of Science (MS)"]},"updated_at":"2026-11-19T07:18:20Z"}