{"id":{"repo_id":"central-wash","oai_identifier":"oai:digitalcommons.cwu.edu:etd-2020"},"canonical_url":"https://search.dev.ndltd.org/etd/central-wash/oai:digitalcommons.cwu.edu:etd-2020","repository":{"repo_id":"central-wash","name":"Central Washington University","base_url":"https://digitalcommons.cwu.edu/do/oai/"},"display":{"title":"Spike-Based Classification of UCI Datasets with Multi-Layer Resume-Like Tempotron","abstract":"Spiking neurons are a class of neuron models that represent information in timed sequences called ``spikes.'' Though predominantly used in neuro-scientific investigations, spiking neural networks (SNN) can be applied to machine learning problems such as classification and regression. SNN are computationally more powerful per neuron than traditional neural networks. Though training time is slow on general purpose computers, spike-based hardware implementations are faster and have shown capability for ultra-low power consumption. Additionally, various SNN training algorithms have achieved comparable performance with the State of the Art on the Fisher Iris dataset. Our main contribution is a software implementation of the multilayer ReSuMe algorithm using the Tempotron principle. The XOR problem is solved in only 13.73 epochs on average. However, training time on four different UCI datasets is slow, and, although decent performance is seen, in most respects the accuracy of our SNN underperforms compared to other SNN, SVM, and ANN experiments. Additionally, our results on the UCI dataset are only preliminary, necessitating further tuning.","abstract_html":"Spiking neurons are a class of neuron models that represent information in timed sequences called ``spikes.'' Though predominantly used in neuro-scientific investigations, spiking neural networks (SNN) can be applied to machine learning problems such as classification and regression. SNN are computationally more powerful per neuron than traditional neural networks. Though training time is slow on general purpose computers, spike-based hardware implementations are faster and have shown capability for ultra-low power consumption. Additionally, various SNN training algorithms have achieved comparable performance with the State of the Art on the Fisher Iris dataset. Our main contribution is a software implementation of the multilayer ReSuMe algorithm using the Tempotron principle. The XOR problem is solved in only 13.73 epochs on average. However, training time on four different UCI datasets is slow, and, although decent performance is seen, in most respects the accuracy of our SNN underperforms compared to other SNN, SVM, and ANN experiments. Additionally, our results on the UCI dataset are only preliminary, necessitating further tuning.","abstract_has_math":false,"creators":["Abdul-Wahid, Sami"],"institution":null,"degree_name":"Master of Science (MS)","degree_level":null,"degree_discipline":"Computational Science","degree_department":null,"school":null,"contributors":["Razvan Andonie","Szilárd Vajda","Donald Davendra"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2018,"date_issued":"2018-01-01T08:00:00Z","date_published":"2018-01-01T08:00:00Z","updated_at":"2026-07-24T01:37:23Z","subjects":["Spiking Neural Networks","Supervised Learning","Iris Dataset","Artificial Intelligence and Robotics","Numerical Analysis and Scientific Computing"],"languages":["English"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.cwu.edu/etd/1008","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Razvan Andonie","Szilárd Vajda","Donald Davendra"]},{"key":"dc:creator","label":"Author","values":["Abdul-Wahid, Sami"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2018-09-17T07:00:00Z"]},{"key":"dc:type","label":"Dc Type","values":["Text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computational Science"]},{"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":["Spiking Neural Networks","Supervised Learning","Iris Dataset","Artificial Intelligence and Robotics","Numerical Analysis and Scientific Computing"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["English"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.cwu.edu/etd/1008"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Spiking neurons are a class of neuron models that represent information in timed sequences called ``spikes.'' Though predominantly used in neuro-scientific investigations, spiking neural networks (SNN) can be applied to machine learning problems such as classification and regression. SNN are computationally more powerful per neuron than traditional neural networks. Though training time is slow on general purpose computers, spike-based hardware implementations are faster and have shown capability for ultra-low power consumption. Additionally, various SNN training algorithms have achieved comparable performance with the State of the Art on the Fisher Iris dataset. Our main contribution is a software implementation of the multilayer ReSuMe algorithm using the Tempotron principle. The XOR problem is solved in only 13.73 epochs on average. However, training time on four different UCI datasets is slow, and, although decent performance is seen, in most respects the accuracy of our SNN underperforms compared to other SNN, SVM, and ANN experiments. Additionally, our results on the UCI dataset are only preliminary, necessitating further tuning."]},{"key":"dc:title","label":"Title","values":["Spike-Based Classification of UCI Datasets with Multi-Layer Resume-Like Tempotron"]}]}],"canonical_facts":{"dc:contributor":["Razvan Andonie","Szilárd Vajda","Donald Davendra"],"dc:creator":["Abdul-Wahid, Sami"],"dc:date.available":["2018-09-17T07:00:00Z"],"dc:description.abstract":["Spiking neurons are a class of neuron models that represent information in timed sequences called ``spikes.'' Though predominantly used in neuro-scientific investigations, spiking neural networks (SNN) can be applied to machine learning problems such as classification and regression. SNN are computationally more powerful per neuron than traditional neural networks. Though training time is slow on general purpose computers, spike-based hardware implementations are faster and have shown capability for ultra-low power consumption. Additionally, various SNN training algorithms have achieved comparable performance with the State of the Art on the Fisher Iris dataset. Our main contribution is a software implementation of the multilayer ReSuMe algorithm using the Tempotron principle. The XOR problem is solved in only 13.73 epochs on average. However, training time on four different UCI datasets is slow, and, although decent performance is seen, in most respects the accuracy of our SNN underperforms compared to other SNN, SVM, and ANN experiments. Additionally, our results on the UCI dataset are only preliminary, necessitating further tuning."],"dc:identifier":["https://digitalcommons.cwu.edu/etd/1008"],"dc:language":["English"],"dc:subject":["Spiking Neural Networks","Supervised Learning","Iris Dataset","Artificial Intelligence and Robotics","Numerical Analysis and Scientific Computing"],"dc:title":["Spike-Based Classification of UCI Datasets with Multi-Layer Resume-Like Tempotron"],"dc:type":["Text"],"thesis:degree_discipline":["Computational Science"],"thesis:degree_name":["Master of Science (MS)"]},"updated_at":"2026-07-24T01:37:23Z"}