Show simple item record

dc.contributor.authorComiter, Marcus Zacharyen_US
dc.date.accessioned2015-07-16T16:26:21Z
dc.date.created2015-05en_US
dc.date.issued2015-06-26en_US
dc.date.submitted2015en_US
dc.identifier.citationComiter, Marcus Zachary. 2015. A Future of Abundant Sparsity: Novel Use and Analysis of Sparse Coding in Machine Learning Applications. Bachelor's thesis, Harvard College.en_US
dc.identifier.urihttp://nrs.harvard.edu/urn-3:HUL.InstRepos:17417575
dc.description.abstractWe present novel applications and analysis of the use of sparse coding within the con- text of machine learning. We first present Sparse Coding Trees (SC-trees), a sparse coding-based framework for resolving classification conflicts, which occur when different classes are mapped to similar feature representations. More specifically, SC-trees are novel supervised hierarchical clustering trees that use node specific dictionary and classifier training to direct input images based on classification results in the feature space at each node. We validate SC-trees on image-based emotion classification, combining it with Mirrored Nonnegative Sparse Coding (MNNSC), a novel sparse coding algorithm leveraging a nonnegativity constraint and the inherent symmetry of the domain, to achieve results exceeding or competitive with the state-of-the-art. We next present SILQ, a sparse coding-based link state model that can predictively buffer packets during wireless link outages to avoid disruption to higher layer protocols such as TCP. We demonstrate empirically that SILQ increases TCP throughput by a factor of 2-4x in varied scenarios.en_US
dc.description.sponsorshipComputer Scienceen_US
dc.format.mimetypeapplication/pdfen_US
dc.language.isoenen_US
dash.licenseMETA_ONLYen_US
dc.subjectComputer Scienceen_US
dc.subjectStatisticsen_US
dc.titleA Future of Abundant Sparsity: Novel Use and Analysis of Sparse Coding in Machine Learning Applicationsen_US
dc.typeThesis or Dissertationen_US
dash.depositing.authorComiter, Marcus Zacharyen_US
dash.embargo.until10000-01-01
thesis.degree.date2015en_US
thesis.degree.grantorHarvard Collegeen_US
thesis.degree.levelUndergraduateen_US
thesis.degree.nameABen_US
dc.type.materialtexten_US
thesis.degree.departmentComputer Scienceen_US
dash.identifier.vireohttp://etds.lib.harvard.edu/college/admin/view/58en_US
dash.title.page1en_US
dash.author.emailmarcuscomiter@gmail.comen_US
thesis.degree.department-secondaryStatisticsen_US
dash.identifier.drsurn-3:HUL.DRS.OBJECT:25267863en_US
dash.identifier.orcid0000-0002-5128-3077en_US
dash.contributor.affiliatedComiter, Marcus
dc.identifier.orcid0000-0002-5128-3077


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record