2026-09-19T09:02:36Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/99116 2023-07-11 col_2142_5131 col_2142_10761 com_2142_5130 com_2142_10755 com_2142_234

Made available in DSpace on 2018-03-02T19:59:44Z (GMT). No. of bitstreams: 2 PHADKE-THESIS-2017.pdf: 14510299 bytes, checksum: e357657487b88e1a0ee47863e6b7f0bc (MD5) LICENSE.txt: 4210 bytes, checksum: fc35d28054217b84ccd68356449c9d17 (MD5) Previous issue date: 2017-07-18 Vaidya, Nitin H. Phadke, Nishad Ashok 2018-03-02T19:59:44Z 2018-03-02T19:59:44Z 2020-03-03T10:15:35Z 2017-07-18 2017-08 Large-scale machine learning has recently risen to prominence in settings of both industry and academia, driven by today's newfound accessibility to data-collecting sensors and high-volume data storage devices. The advent of these capabilities in industry, however, has raised questions about the privacy implications of new massively data-driven, subscribable services offered by corporations to individuals. Recent lines of research have developed algorithms designed to scale in distributed machine learning environments that make certain privacy guarantees to subscribers without hindering the quality of service the corporations are able to provide. In this work, we fully implement one such distributed optimization framework and rigorously test its parameterized convergence properties. We also develop a system of both disruptive and nondisruptive attacks designed to aggressively intrude upon subscribers' privacy and to glean subscribers' private data from information readily available within the framework's network. These attack techniques can be seamlessly integrated into the aforementioned distributed optimization framework and are shown to be a risk to the privacy of the system. Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2019-08-01 The student, Nishad Phadke, accepted the attached license on 2017-07-18 at 10:39. The student, Nishad Phadke, submitted this Thesis for approval on 2017-07-18 at 10:47. This Thesis was approved for publication on 2017-07-18 at 11:39. DSpace SAF Submission Ingestion Package generated from Vireo submission #11509 on 2018-03-02 at 13:02:26 Embargo set by: Seth Robbins for item 105070 Lift date: 2020-03-02T19:59:52Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system Embargo set by: Seth Robbins for item 105070 Lift date: 2020-03-02T20:02:46Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system U of I Only Restriction Lifted for Item 105070 on 2020-03-03T10:15:35Z. application/pdf http://hdl.handle.net/2142/99116 en Copyright 2017 Nishad Phadke Distributed optimization Privacy A framework for privacy-preserving, distributed machine learning using gradient obfuscation text text Computer Science Computer Science University of Illinois at Urbana-Champaign Thesis M.S.