Formal Privacy for Functional Data with Gaussian Perturbations

Motivated by the rapid rise in statistical tools in Functional Data Analysis, we consider the Gaussian mechanism for achieving differential privacy with parameter estimates taking values in a, potentially infinite-dimensional, separable Banach space. Using classic results from probability theory, we...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Mirshani, Ardalan, Reimherr, Matthew, Slavkovic, Aleksandra
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Mirshani, Ardalan
Reimherr, Matthew
Slavkovic, Aleksandra
description Motivated by the rapid rise in statistical tools in Functional Data Analysis, we consider the Gaussian mechanism for achieving differential privacy with parameter estimates taking values in a, potentially infinite-dimensional, separable Banach space. Using classic results from probability theory, we show how densities over function spaces can be utilized to achieve the desired differential privacy bounds. This extends prior results of Hall et al (2013) to a much broader class of statistical estimates and summaries, including "path level" summaries, nonlinear functionals, and full function releases. By focusing on Banach spaces, we provide a deeper picture of the challenges for privacy with complex data, especially the role regularization plays in balancing utility and privacy. Using an application to penalized smoothing, we explicitly highlight this balance in the context of mean function estimation. Simulations and an application to diffusion tensor imaging are briefly presented, with extensive additions included in a supplement.
doi_str_mv 10.48550/arxiv.1711.06660
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1711_06660</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1711_06660</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-a46f514c0f666d8c32065237226511f33cbc6c65947c05d6abf3b69779d07b6b3</originalsourceid><addsrcrecordid>eNotj8FugzAQRH3JoSL9gJ7qH4CuMV7DJVKVFloJKTnkjtYGq5YSiAyk5e-bpD2NNBq90WPsSUCS5UrBC4Uff0mEFiIBRIQHtimHcKIj3wd_IbtwNwRezr2d_NBf6zeaiH_76YtXNI-jp57vuzDNwdBtMa7ZytFx7B7_M2KH8v2w_YjrXfW5fa1jQg0xZeiUyCy462mbW5kCqlTqNEUlhJPSGosWVZFpC6pFMk4aLLQuWtAGjYzY8x_2LtCcgz9RWJqbSHMXkb86A0Jc</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Formal Privacy for Functional Data with Gaussian Perturbations</title><source>arXiv.org</source><creator>Mirshani, Ardalan ; Reimherr, Matthew ; Slavkovic, Aleksandra</creator><creatorcontrib>Mirshani, Ardalan ; Reimherr, Matthew ; Slavkovic, Aleksandra</creatorcontrib><description>Motivated by the rapid rise in statistical tools in Functional Data Analysis, we consider the Gaussian mechanism for achieving differential privacy with parameter estimates taking values in a, potentially infinite-dimensional, separable Banach space. Using classic results from probability theory, we show how densities over function spaces can be utilized to achieve the desired differential privacy bounds. This extends prior results of Hall et al (2013) to a much broader class of statistical estimates and summaries, including "path level" summaries, nonlinear functionals, and full function releases. By focusing on Banach spaces, we provide a deeper picture of the challenges for privacy with complex data, especially the role regularization plays in balancing utility and privacy. Using an application to penalized smoothing, we explicitly highlight this balance in the context of mean function estimation. Simulations and an application to diffusion tensor imaging are briefly presented, with extensive additions included in a supplement.</description><identifier>DOI: 10.48550/arxiv.1711.06660</identifier><language>eng</language><subject>Mathematics - Statistics Theory ; Statistics - Theory</subject><creationdate>2017-11</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1711.06660$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1711.06660$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Mirshani, Ardalan</creatorcontrib><creatorcontrib>Reimherr, Matthew</creatorcontrib><creatorcontrib>Slavkovic, Aleksandra</creatorcontrib><title>Formal Privacy for Functional Data with Gaussian Perturbations</title><description>Motivated by the rapid rise in statistical tools in Functional Data Analysis, we consider the Gaussian mechanism for achieving differential privacy with parameter estimates taking values in a, potentially infinite-dimensional, separable Banach space. Using classic results from probability theory, we show how densities over function spaces can be utilized to achieve the desired differential privacy bounds. This extends prior results of Hall et al (2013) to a much broader class of statistical estimates and summaries, including "path level" summaries, nonlinear functionals, and full function releases. By focusing on Banach spaces, we provide a deeper picture of the challenges for privacy with complex data, especially the role regularization plays in balancing utility and privacy. Using an application to penalized smoothing, we explicitly highlight this balance in the context of mean function estimation. Simulations and an application to diffusion tensor imaging are briefly presented, with extensive additions included in a supplement.</description><subject>Mathematics - Statistics Theory</subject><subject>Statistics - Theory</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2017</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj8FugzAQRH3JoSL9gJ7qH4CuMV7DJVKVFloJKTnkjtYGq5YSiAyk5e-bpD2NNBq90WPsSUCS5UrBC4Uff0mEFiIBRIQHtimHcKIj3wd_IbtwNwRezr2d_NBf6zeaiH_76YtXNI-jp57vuzDNwdBtMa7ZytFx7B7_M2KH8v2w_YjrXfW5fa1jQg0xZeiUyCy462mbW5kCqlTqNEUlhJPSGosWVZFpC6pFMk4aLLQuWtAGjYzY8x_2LtCcgz9RWJqbSHMXkb86A0Jc</recordid><startdate>20171117</startdate><enddate>20171117</enddate><creator>Mirshani, Ardalan</creator><creator>Reimherr, Matthew</creator><creator>Slavkovic, Aleksandra</creator><scope>AKZ</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20171117</creationdate><title>Formal Privacy for Functional Data with Gaussian Perturbations</title><author>Mirshani, Ardalan ; Reimherr, Matthew ; Slavkovic, Aleksandra</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-a46f514c0f666d8c32065237226511f33cbc6c65947c05d6abf3b69779d07b6b3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2017</creationdate><topic>Mathematics - Statistics Theory</topic><topic>Statistics - Theory</topic><toplevel>online_resources</toplevel><creatorcontrib>Mirshani, Ardalan</creatorcontrib><creatorcontrib>Reimherr, Matthew</creatorcontrib><creatorcontrib>Slavkovic, Aleksandra</creatorcontrib><collection>arXiv Mathematics</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Mirshani, Ardalan</au><au>Reimherr, Matthew</au><au>Slavkovic, Aleksandra</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Formal Privacy for Functional Data with Gaussian Perturbations</atitle><date>2017-11-17</date><risdate>2017</risdate><abstract>Motivated by the rapid rise in statistical tools in Functional Data Analysis, we consider the Gaussian mechanism for achieving differential privacy with parameter estimates taking values in a, potentially infinite-dimensional, separable Banach space. Using classic results from probability theory, we show how densities over function spaces can be utilized to achieve the desired differential privacy bounds. This extends prior results of Hall et al (2013) to a much broader class of statistical estimates and summaries, including "path level" summaries, nonlinear functionals, and full function releases. By focusing on Banach spaces, we provide a deeper picture of the challenges for privacy with complex data, especially the role regularization plays in balancing utility and privacy. Using an application to penalized smoothing, we explicitly highlight this balance in the context of mean function estimation. Simulations and an application to diffusion tensor imaging are briefly presented, with extensive additions included in a supplement.</abstract><doi>10.48550/arxiv.1711.06660</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1711.06660
ispartof
issn
language eng
recordid cdi_arxiv_primary_1711_06660
source arXiv.org
subjects Mathematics - Statistics Theory
Statistics - Theory
title Formal Privacy for Functional Data with Gaussian Perturbations
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-01T13%3A21%3A13IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Formal%20Privacy%20for%20Functional%20Data%20with%20Gaussian%20Perturbations&rft.au=Mirshani,%20Ardalan&rft.date=2017-11-17&rft_id=info:doi/10.48550/arxiv.1711.06660&rft_dat=%3Carxiv_GOX%3E1711_06660%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true