Show simple item record

dc.contributor.authorGao, Yuguang
dc.date.accessioned2019-10-15T16:49:21Z
dc.date.issued2019-08-30
dc.identifier.otherGao_cornellgrad_0058F_11340
dc.identifier.otherhttp://dissertations.umi.com/cornellgrad:11340
dc.identifier.otherbibid: 11050641
dc.identifier.urihttps://hdl.handle.net/1813/67657
dc.description.abstractThis thesis discusses a few interesting topics regarding fundamental aspects of learning in the following prevalent application scenarios: (1) training a neural network for image classification, and (2) using a channel code for universal communication at capacity. For the classification problem, we aim to develop a better understanding of what representations each layer of the network learns. In particular, we compare the higher-layer activations of two neural networks with identical architecture but different initializations via adaptive $k$NN graph approximation of the underlying manifold, and we show that there are vast similarities between the underlying manifolds of the two networks but with discrepancy in potentially highly-curved regions. We also investigate locality of the receptive field in the Convolutional Neural Networks by using semi-localized filters with random neuron connection, where we find out that the receptive field might be beyond local for feature extraction as is hard coded in traditional design. For the communication problem, we study universal channel coding under the high-dimensional statistical setting beyond Shannon's classical framework, and we prove a series of theorems that may surprisingly indicate a need to learn the entire channel in order to achieve its capacity.
dc.language.isoen_US
dc.subjectElectrical engineering
dc.titleOn Some Fundamental Aspects of Learning in Artificial Neural Networks and Universal Channel Codes
dc.typedissertation or thesis
dc.description.embargo2021-08-29
thesis.degree.disciplineElectrical and Computer Engineering
thesis.degree.grantorCornell University
thesis.degree.levelDoctor of Philosophy
thesis.degree.namePh.D., Electrical and Computer Engineering
dc.contributor.chairHopcroft, John E.
dc.contributor.committeeMemberTong, Lang
dc.contributor.committeeMemberKleinberg, Robert David
dcterms.licensehttps://hdl.handle.net/1813/59810
dc.identifier.doihttps://doi.org/10.7298/6a79-d633


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

Statistics