Aggregation for Sensitive Data

Avradeep Bhowmik, Joydeep Ghosh, Oluwasanmi Koyejo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In many modern applications, considerations like privacy, security and legal doctrines like the GDPR put limitations on data storage and sharing with third parties. Specifically, access to individual level data points is restricted and machine learning models need to be trained with aggregated versions of the datasets. Learning with aggregated data is a new and relatively unexplored form of semi-supervision. We tackle this problem by designing aggregation paradigms that conform to certain kinds of privacy or non-identifiability requirements. We further develop novel learning algorithms that can nevertheless be used to learn from only these aggregates. We motivate our framework for the case of Gaussian regression, and subsequently extend our techniques to subsume arbitrary binary classifiers and generalised linear models. We provide theoretical results and empirical evaluation of our methods on real data from healthcare and telecom.

Original languageEnglish (US)
Title of host publication2019 13th International Conference on Sampling Theory and Applications, SampTA 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728137414
DOIs
StatePublished - Jul 2019
Event13th International Conference on Sampling Theory and Applications, SampTA 2019 - Bordeaux, France
Duration: Jul 8 2019Jul 12 2019

Publication series

Name2019 13th International Conference on Sampling Theory and Applications, SampTA 2019

Conference

Conference13th International Conference on Sampling Theory and Applications, SampTA 2019
CountryFrance
CityBordeaux
Period7/8/197/12/19

ASJC Scopus subject areas

  • Statistics and Probability
  • Signal Processing
  • Analysis
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Aggregation for Sensitive Data'. Together they form a unique fingerprint.

Cite this