Show simple item record

dc.contributor.advisorGray, Alexander
dc.contributor.authorGanti Mahapatruni, Ravi Sastry
dc.date.accessioned2014-05-22T15:23:35Z
dc.date.available2014-05-22T15:23:35Z
dc.date.created2014-05
dc.date.issued2014-01-10
dc.date.submittedMay 2014
dc.identifier.urihttp://hdl.handle.net/1853/51801
dc.description.abstractIn this thesis, we provide computationally efficient algorithms with provable statistical guarantees, for the problem of active learning, by using ideas from sequential analysis. We provide a generic algorithmic framework for active learning in the pool setting, and instantiate this framework by using ideas from learning with experts, stochastic optimization, and multi-armed bandits. For the problem of learning convex combination of a given set of hypothesis, we provide a stochastic mirror descent based active learning algorithm in the stream setting.
dc.format.mimetypeapplication/pdf
dc.language.isoen_US
dc.publisherGeorgia Institute of Technology
dc.subjectActive learning
dc.subjectSequential analysis
dc.subjectStochastic optimization
dc.subject.lcshActive learning
dc.subject.lcshAlgorithms
dc.subject.lcshSequential analysis
dc.subject.lcshMathematical optimization
dc.subject.lcshMachine learning
dc.titleNew formulations for active learning
dc.typeDissertation
dc.description.degreePh.D.
dc.contributor.departmentComputer Science
thesis.degree.levelDoctoral
dc.contributor.committeeMemberBalcan, Maria-Florina
dc.contributor.committeeMemberSong, Le
dc.contributor.committeeMemberRakhlin, Alexander
dc.contributor.committeeMemberZhang, Tong
dc.date.updated2014-05-22T15:23:35Z


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record