List of implementations of differentially private analyses

{{redirect-distinguish|Pinq|Pink (disambiguation){{!}}Pink|Pinque (disambiguation){{!}}Pinque}}{{Notability|Lists|date=July 2024}}

Since the advent of differential privacy, a number of systems supporting differentially private data analyses have been implemented and deployed. This article tracks real-world deployments, production software packages, and research prototypes.

Real-world deployments

class="sortable wikitable" style="text-align: left"
Name ! Organization ! Year Introduced ! Notes ! Still in use?
OnTheMap: Interactive tool for exploration of US income and commute patterns.{{Cite web\|url=https://onthemap.ces.census.gov/\|title=OnTheMap\|website=onthemap.ces.census.gov\|accessdate=29 March 2023}}{{cite book \|last1=Machanavajjhala \|first1=Ashwin \|last2=Kifer \|first2=Daniel \|last3=Abowd \|first3=John \|last4=Gehrke \|first4=Johannes \|last5=Vilhuber \|first5=Lars \|title=2008 IEEE 24th International Conference on Data Engineering \|chapter=Privacy: Theory meets Practice on the Map \|pages=277–286 \|date=April 2008 \|doi=10.1109/ICDE.2008.4497436\|isbn=978-1-4244-1836-7 \|s2cid=5812674 }} \| US Census Bureau \| 2008 \| First deployment of differential privacy \| Yes
RAPPOR in Chrome Browser to collect security metrics{{cite web \|last1=Erlingsson \|first1=Úlfar \|title=Learning statistics with privacy, aided by the flip of a coin \|url=https://security.googleblog.com/2014/10/learning-statistics-with-privacy-aided.html}}{{cite book \|last1=Erlingsson \|first1=Úlfar \|last2=Pihur \|first2=Vasyl \|last3=Korolova \|first3=Aleksandra \|title=Proceedings of the 2014 ACM SIGSAC Conference on Computer and Communications Security \|chapter=RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response \|url=https://archive.org/details/arxiv-1407.6981 \|date=November 2014 \|pages=1054–1067 \|doi=10.1145/2660267.2660348\|bibcode=2014arXiv1407.6981E \|arxiv=1407.6981 \|isbn=9781450329576 \|s2cid=6855746 }} \|Google \|2014 \|First widespread use of local differential privacy \|No
Emoji analytics; analytics. Improve: QuickType, emoji; Spotlight deep link suggestions; Lookup Hints in Notes. Emoji suggestions, health type usage estimates, Safari energy drain statistics, Autoplay intent detection (also in Safari){{cite journal\|last1=Differential Privacy Team \|title=Learning with Privacy at Scale \|journal=Apple Machine Learning Journal \|date=December 2017 \|volume=1 \|issue=8 \|url=https://machinelearning.apple.com/2017/12/06/learning-with-privacy-at-scale.html}} \| Apple \| 2017 \| \| Yes
Application telemetry{{cite journal \|last1=Ding \|first1=Bolin \|last2=Kulkarni \|first2=Janardhan \|last3=Yekhanin \|first3=Sergey \|title=Collecting Telemetry Data Privately \|journal=31st Conference on Neural Information Processing Systems \|date=December 2017 \|pages=3574–3583\|bibcode=2017arXiv171201524D \|arxiv=1712.01524 }} \| Microsoft \| 2017 \| Application usage statistics Microsoft Windows. \| yes
Flex: A SQL-based system developed for internal Uber analytics{{cite web \|last1=Tezapsidis \|first1=Katie \|title=Uber Releases Open Source Project for Differential Privacy \|url=https://medium.com/uber-security-privacy/differential-privacy-open-source-7892c82c42b6 \|date=Jul 13, 2017}}{{cite journal \|last1=Johnson \|first1=Noah \|last2=Near \|first2=Joseph P. \|last3=Song \|first3=Dawn \|title=Towards Practical Differential Privacy for SQL Queries \|journal=Proceedings of the VLDB Endowment \|date=January 2018 \|volume=11 \|issue=5 \|pages=526–539 \|doi=10.1145/3187009.3177733\|arxiv=1706.09479 }} \| Uber \| 2017 \| \| Unknown
2020 Census{{cite book \|last1=Abowd \|first1=John M. \|title=Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining \|chapter=The U.S. Census Bureau Adopts Differential Privacy \|date=August 2018 \|page=2867 \|doi=10.1145/3219819.3226070\|isbn=9781450355520 \|chapter-url=https://digitalcommons.ilr.cornell.edu/ldi/49 \|hdl=1813/60392 \|s2cid=51711121 \|hdl-access=free }} \|US Census Bureau \| 2018 \| \| Yes
Audience Engagement API{{Cite arXiv\| eprint=2002.05839 \| last1=Rogers \| first1=Ryan \| last2=Subramaniam \| first2=Subbu \| last3=Peng \| first3=Sean \| last4=Durfee \| first4=David \| last5=Lee \| first5=Seunghyun \| author6=Santosh Kumar Kancha \| last7=Sahay \| first7=Shraddha \| last8=Ahammad \| first8=Parvez \| title=LinkedIn's Audience Engagements API: A Privacy Preserving Data Analytics System at Scale \| year=2020 \| class=cs.CR }} \|LinkedIn \|2020 \| \|Yes
Labor Market Insights{{Cite arXiv\| eprint=2010.13981 \| last1=Rogers \| first1=Ryan \| author2=Adrian Rivera Cardoso \| last3=Mancuhan \| first3=Koray \| last4=Kaura \| first4=Akash \| last5=Gahlawat \| first5=Nikhil \| last6=Jain \| first6=Neha \| last7=Ko \| first7=Paul \| last8=Ahammad \| first8=Parvez \| title=A Members First Approach to Enabling LinkedIn's Labor Market Insights at Scale \| year=2020 \| class=cs.CR }} \|LinkedIn \|2020 \| \|Yes
COVID-19 Community Mobility Reports{{Cite arXiv\| eprint=2004.04145 \| last1=Aktay \| first1=Ahmet \| last2=Bavadekar \| first2=Shailesh \| last3=Cossoul \| first3=Gwen \| last4=Davis \| first4=John \| last5=Desfontaines \| first5=Damien \| last6=Fabrikant \| first6=Alex \| last7=Gabrilovich \| first7=Evgeniy \| last8=Gadepalli \| first8=Krishna \| last9=Gipson \| first9=Bryant \| last10=Guevara \| first10=Miguel \| last11=Kamath \| first11=Chaitanya \| last12=Kansal \| first12=Mansi \| last13=Lange \| first13=Ali \| last14=Mandayam \| first14=Chinmoy \| last15=Oplinger \| first15=Andrew \| last16=Pluntke \| first16=Christopher \| last17=Roessler \| first17=Thomas \| last18=Schlosberg \| first18=Arran \| last19=Shekel \| first19=Tomer \| last20=Vispute \| first20=Swapnil \| last21=Vu \| first21=Mia \| last22=Wellenius \| first22=Gregory \| last23=Williams \| first23=Brian \| author24=Royce J Wilson \| title=Google COVID-19 Community Mobility Reports: Anonymization Process Description (Version 1.1) \| year=2020 \| class=cs.CR }} \|Google \|2020 \| \|Unknown
Advertiser Queries{{Cite arXiv\|eprint=2002.05839\|last1=Rogers\|first1=Ryan\|last2=Subbu\|first2=Subramaniam\|first3=Sean\|last3=Peng\|first4=David\|last4=Durfee\|first5=Seunghyun\|last5=Lee\|first6=Santosh Kumar\|last6=Kancha\|first7=Shraddha\|last7=Sahay\|first8=Parvez\|last8=Ahammad\|title=LinkedIn's Audience Engagements API: A Privacy Preserving Data Analytics System at Scale\|year=2020\|class=cs.CR }} \|LinkedIn \|2020 \|
U.S. Broadband Coverage Data Set{{Cite arXiv\| eprint=2103.14035 \| last1=Pereira \| first1=Mayana \| last2=Kim \| first2=Allen \| last3=Allen \| first3=Joshua \| last4=White \| first4=Kevin \| author5=Juan Lavista Ferres \| last6=Dodhia \| first6=Rahul \| title=U.S. Broadband Coverage Data Set: A Differentially Private Data Release \| year=2021 \| class=cs.CR }} \|Microsoft \|2021 \| \|Unknown
[https://collegescorecard.ed.gov/ College Scorecard Website] \|IRS and Dept. of Education \|2021 \| \|Unknown
Ohm Connect{{Cite web\|url=https://edp.recurve.com/\|title=EDP\|website=EDP\|accessdate=29 March 2023}} \|Recurve \|2021 \| \|
[https://birth.dataset.pub/ Live Birth Dataset]{{Cite arXiv\|eprint= 2405.00267\|last1=Hod\|first1=Shlomi\|last2=Canetti\|first2=Ran\|title=Differentially Private Release of Israel's National Registry of Live Births\|year=2024\|class=cs.CR }}{{Cite web\|url=https://data.gov.il/dataset/birth-data\|title=Live Birth Dataset (Hebrew)\|website=data.gov.il\|accessdate=2 May 2024}} \|Israeli Ministry of Health \|2024 \| \|Yes

class="sortable wikitable" style="text-align: left"

Name

! Organization

! Year Introduced

! Notes

! Still in use?

OnTheMap: Interactive tool for exploration of US income and commute patterns.{{Cite web|url=https://onthemap.ces.census.gov/|title=OnTheMap|website=onthemap.ces.census.gov|accessdate=29 March 2023}}{{cite book |last1=Machanavajjhala |first1=Ashwin |last2=Kifer |first2=Daniel |last3=Abowd |first3=John |last4=Gehrke |first4=Johannes |last5=Vilhuber |first5=Lars |title=2008 IEEE 24th International Conference on Data Engineering |chapter=Privacy: Theory meets Practice on the Map |pages=277–286 |date=April 2008 |doi=10.1109/ICDE.2008.4497436|isbn=978-1-4244-1836-7 |s2cid=5812674 }}

| US Census Bureau

| 2008

| First deployment of differential privacy

| Yes

RAPPOR in Chrome Browser to collect security metrics{{cite web |last1=Erlingsson |first1=Úlfar |title=Learning statistics with privacy, aided by the flip of a coin |url=https://security.googleblog.com/2014/10/learning-statistics-with-privacy-aided.html}}{{cite book |last1=Erlingsson |first1=Úlfar |last2=Pihur |first2=Vasyl |last3=Korolova |first3=Aleksandra |title=Proceedings of the 2014 ACM SIGSAC Conference on Computer and Communications Security |chapter=RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response |url=https://archive.org/details/arxiv-1407.6981 |date=November 2014 |pages=1054–1067 |doi=10.1145/2660267.2660348|bibcode=2014arXiv1407.6981E |arxiv=1407.6981 |isbn=9781450329576 |s2cid=6855746 }}

|Google

|2014

|First widespread use of local differential privacy

|No

Emoji analytics; analytics. Improve: QuickType, emoji; Spotlight deep link suggestions; Lookup Hints in Notes. Emoji suggestions, health type usage estimates, Safari energy drain statistics, Autoplay intent detection (also in Safari){{cite journal|last1=Differential Privacy Team |title=Learning with Privacy at Scale |journal=Apple Machine Learning Journal |date=December 2017 |volume=1 |issue=8 |url=https://machinelearning.apple.com/2017/12/06/learning-with-privacy-at-scale.html}}

| Apple

| 2017

| Yes

Application telemetry{{cite journal |last1=Ding |first1=Bolin |last2=Kulkarni |first2=Janardhan |last3=Yekhanin |first3=Sergey |title=Collecting Telemetry Data Privately |journal=31st Conference on Neural Information Processing Systems |date=December 2017 |pages=3574–3583|bibcode=2017arXiv171201524D |arxiv=1712.01524 }}

| Microsoft

| 2017

| Application usage statistics Microsoft Windows.

| yes

Flex: A SQL-based system developed for internal Uber analytics{{cite web |last1=Tezapsidis |first1=Katie |title=Uber Releases Open Source Project for Differential Privacy |url=https://medium.com/uber-security-privacy/differential-privacy-open-source-7892c82c42b6 |date=Jul 13, 2017}}{{cite journal |last1=Johnson |first1=Noah |last2=Near |first2=Joseph P. |last3=Song |first3=Dawn |title=Towards Practical Differential Privacy for SQL Queries |journal=Proceedings of the VLDB Endowment |date=January 2018 |volume=11 |issue=5 |pages=526–539 |doi=10.1145/3187009.3177733|arxiv=1706.09479 }}

| Uber

| 2017

| Unknown

2020 Census{{cite book |last1=Abowd |first1=John M. |title=Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining |chapter=The U.S. Census Bureau Adopts Differential Privacy |date=August 2018 |page=2867 |doi=10.1145/3219819.3226070|isbn=9781450355520 |chapter-url=https://digitalcommons.ilr.cornell.edu/ldi/49 |hdl=1813/60392 |s2cid=51711121 |hdl-access=free }}

|US Census Bureau

| 2018

| Yes

Audience Engagement API{{Cite arXiv| eprint=2002.05839 | last1=Rogers | first1=Ryan | last2=Subramaniam | first2=Subbu | last3=Peng | first3=Sean | last4=Durfee | first4=David | last5=Lee | first5=Seunghyun | author6=Santosh Kumar Kancha | last7=Sahay | first7=Shraddha | last8=Ahammad | first8=Parvez | title=LinkedIn's Audience Engagements API: A Privacy Preserving Data Analytics System at Scale | year=2020 | class=cs.CR }}

|LinkedIn

|2020

|Yes

Labor Market Insights{{Cite arXiv| eprint=2010.13981 | last1=Rogers | first1=Ryan | author2=Adrian Rivera Cardoso | last3=Mancuhan | first3=Koray | last4=Kaura | first4=Akash | last5=Gahlawat | first5=Nikhil | last6=Jain | first6=Neha | last7=Ko | first7=Paul | last8=Ahammad | first8=Parvez | title=A Members First Approach to Enabling LinkedIn's Labor Market Insights at Scale | year=2020 | class=cs.CR }}

|LinkedIn

|2020

|Yes

COVID-19 Community Mobility Reports{{Cite arXiv| eprint=2004.04145 | last1=Aktay | first1=Ahmet | last2=Bavadekar | first2=Shailesh | last3=Cossoul | first3=Gwen | last4=Davis | first4=John | last5=Desfontaines | first5=Damien | last6=Fabrikant | first6=Alex | last7=Gabrilovich | first7=Evgeniy | last8=Gadepalli | first8=Krishna | last9=Gipson | first9=Bryant | last10=Guevara | first10=Miguel | last11=Kamath | first11=Chaitanya | last12=Kansal | first12=Mansi | last13=Lange | first13=Ali | last14=Mandayam | first14=Chinmoy | last15=Oplinger | first15=Andrew | last16=Pluntke | first16=Christopher | last17=Roessler | first17=Thomas | last18=Schlosberg | first18=Arran | last19=Shekel | first19=Tomer | last20=Vispute | first20=Swapnil | last21=Vu | first21=Mia | last22=Wellenius | first22=Gregory | last23=Williams | first23=Brian | author24=Royce J Wilson | title=Google COVID-19 Community Mobility Reports: Anonymization Process Description (Version 1.1) | year=2020 | class=cs.CR }}

|Google

|2020

|Unknown

Advertiser Queries{{Cite arXiv|eprint=2002.05839|last1=Rogers|first1=Ryan|last2=Subbu|first2=Subramaniam|first3=Sean|last3=Peng|first4=David|last4=Durfee|first5=Seunghyun|last5=Lee|first6=Santosh Kumar|last6=Kancha|first7=Shraddha|last7=Sahay|first8=Parvez|last8=Ahammad|title=LinkedIn's Audience Engagements API: A Privacy Preserving Data Analytics System at Scale|year=2020|class=cs.CR }}

|LinkedIn

|2020

U.S. Broadband Coverage Data Set{{Cite arXiv| eprint=2103.14035 | last1=Pereira | first1=Mayana | last2=Kim | first2=Allen | last3=Allen | first3=Joshua | last4=White | first4=Kevin | author5=Juan Lavista Ferres | last6=Dodhia | first6=Rahul | title=U.S. Broadband Coverage Data Set: A Differentially Private Data Release | year=2021 | class=cs.CR }}

|Microsoft

|2021

|Unknown

[https://collegescorecard.ed.gov/ College Scorecard Website]

|IRS and Dept. of Education

|2021

|Unknown

Ohm Connect{{Cite web|url=https://edp.recurve.com/|title=EDP|website=EDP|accessdate=29 March 2023}}

|Recurve

|2021

[https://birth.dataset.pub/ Live Birth Dataset]{{Cite arXiv|eprint= 2405.00267|last1=Hod|first1=Shlomi|last2=Canetti|first2=Ran|title=Differentially Private Release of Israel's National Registry of Live Births|year=2024|class=cs.CR }}{{Cite web|url=https://data.gov.il/dataset/birth-data|title=Live Birth Dataset (Hebrew)|website=data.gov.il|accessdate=2 May 2024}}

|Israeli Ministry of Health

|2024

|Yes

Production software packages

These software packages purport to be usable in production systems. They are split in two categories: those focused on answering statistical queries with differential privacy, and those focused on training machine learning models with differential privacy.

= Statistical analyses =

class="sortable wikitable" style="text-align: left"
Name ! Developer ! Year Introduced ! Notes ! Still maintained?
Google's differential privacy libraries{{cite web \| url=https://github.com/google/differential-privacy \| title=Google's differential privacy libraries \| website=GitHub \| date=3 February 2023 }} \| Google \| 2019 \| Building block libraries in Go, C++, and Java; end-to-end framework in Go,.{{cite web \| url=https://github.com/google/differential-privacy/tree/main/privacy-on-beam \| title=Differential-privacy/Privacy-on-beam at main · google/Differential-privacy \| website=GitHub }} \| Yes
OpenDP{{Cite web\|url=https://opendp.org/home\|title=OpenDP\|website=opendp.org\|accessdate=29 March 2023}} \|Harvard, Microsoft \|2020 \|Core library in Rust,{{cite web \| url=https://github.com/opendp/opendp \| title=OpenDP Library \| website=GitHub }} SDK in Python with an SQL interface. \|Yes
Tumult Analytics{{Cite web\|url=https://www.tmlt.dev/\|title=Tumult Analytics\|website=www.tmlt.dev\|accessdate=29 March 2023}} \| Tumult Labs{{Cite web\|url=https://www.tmlt.io/\|title=Tumult Labs \| Privacy Protection Redefined\|website=www.tmlt.io\|accessdate=29 March 2023}} \| 2022 \| Python library, running on Apache Spark. \| Yes
PipelineDP{{Cite web\|url=https://pipelinedp.io/\|title=PipelineDP\|website=pipelinedp.io\|accessdate=29 March 2023}} \| Google, OpenMined{{Cite web\|url=https://www.openmined.org/\|title=OpenMined\|website=www.openmined.org\|accessdate=29 March 2023}} \| 2022 \| Python library, running on Apache Spark, Apache Beam, or locally. \| Yes
PSI (Ψ): A Private data Sharing Interface \| Harvard University Privacy Tools Project.{{cite web \|last1=Gaboardi \|first1=Marco \|last2=Honaker \|first2=James \|last3=King \|first3=Gary \|last4=Nissim \|first4=Kobbi \|last5=Ullman \|first5=Jonathan \|last6=Vadhan \|first6=Salil \|last7=Murtagh \|first7=Jack \|title=PSI (Ψ): a Private data Sharing Interface \|url=https://privacytools.seas.harvard.edu/publications/psipaper \|date=June 2016}} \| 2016 \| \| No
TopDown Algorithm{{cite web \| url=https://github.com/uscensusbureau/DAS_2020_Redistricting_Production_Code \| title=DAS 2020 Redistricting Production Code Release \| website=GitHub \| date=22 June 2022 }} \|United States Census Bureau \|2020 \|Production code used in the 2020 US Census. \|No

class="sortable wikitable" style="text-align: left"

Name

! Developer

! Year Introduced

! Notes

! Still maintained?

Google's differential privacy libraries{{cite web | url=https://github.com/google/differential-privacy | title=Google's differential privacy libraries | website=GitHub | date=3 February 2023 }}

| Google

| 2019

| Building block libraries in Go, C++, and Java; end-to-end framework in Go,.{{cite web | url=https://github.com/google/differential-privacy/tree/main/privacy-on-beam | title=Differential-privacy/Privacy-on-beam at main · google/Differential-privacy | website=GitHub }}

| Yes

OpenDP{{Cite web|url=https://opendp.org/home|title=OpenDP|website=opendp.org|accessdate=29 March 2023}}

|Harvard, Microsoft

|2020

|Core library in Rust,{{cite web | url=https://github.com/opendp/opendp | title=OpenDP Library | website=GitHub }} SDK in Python with an SQL interface.

|Yes

Tumult Analytics{{Cite web|url=https://www.tmlt.dev/|title=Tumult Analytics|website=www.tmlt.dev|accessdate=29 March 2023}}

| Tumult Labs{{Cite web|url=https://www.tmlt.io/|title=Tumult Labs | Privacy Protection Redefined|website=www.tmlt.io|accessdate=29 March 2023}}

| 2022

| Python library, running on Apache Spark.

| Yes

PipelineDP{{Cite web|url=https://pipelinedp.io/|title=PipelineDP|website=pipelinedp.io|accessdate=29 March 2023}}

| Google, OpenMined{{Cite web|url=https://www.openmined.org/|title=OpenMined|website=www.openmined.org|accessdate=29 March 2023}}

| 2022

| Python library, running on Apache Spark, Apache Beam, or locally.

| Yes

PSI (Ψ): A Private data Sharing Interface

| Harvard University Privacy Tools Project.{{cite web |last1=Gaboardi |first1=Marco |last2=Honaker |first2=James |last3=King |first3=Gary |last4=Nissim |first4=Kobbi |last5=Ullman |first5=Jonathan |last6=Vadhan |first6=Salil |last7=Murtagh |first7=Jack |title=PSI (Ψ): a Private data Sharing Interface |url=https://privacytools.seas.harvard.edu/publications/psipaper |date=June 2016}}

| 2016

| No

TopDown Algorithm{{cite web | url=https://github.com/uscensusbureau/DAS_2020_Redistricting_Production_Code | title=DAS 2020 Redistricting Production Code Release | website=GitHub | date=22 June 2022 }}

|United States Census Bureau

|2020

|Production code used in the 2020 US Census.

|No

= Machine learning =

class="sortable wikitable" style="text-align: left"
Name ! Developer ! Year Introduced ! Notes ! Still maintained?
Diffprivlib{{cite web \| url=https://github.com/IBM/differential-privacy-library \| title=Diffprivlib v0.5 \| website=GitHub \| date=17 October 2022 }} \|IBM{{Cite arXiv\| eprint=1907.02444 \| last1=Holohan \| first1=Naoise \| last2=Braghin \| first2=Stefano \| author3=Pól Mac Aonghusa \| last4=Levacher \| first4=Killian \| title=Diffprivlib: The IBM Differential Privacy Library \| year=2019 \| class=cs.CR }} \|2019 \|Python library. \|Yes
TensorFlow Privacy{{cite web \|last1=Radebaugh \|first1=Carey \|last2=Erlingsson \|first2=Ulfar \|title=Introducing TensorFlow Privacy: Learning with Differential Privacy for Training Data \|work=Medium \|url=https://medium.com/tensorflow/introducing-tensorflow-privacy-learning-with-differential-privacy-for-training-data-b143c5e801b6 \|date=March 6, 2019}}{{cite web \|title=TensorFlow Privacy \|website=GitHub\|url=https://github.com/tensorflow/privacy\|date=2019-08-09}} \|Google \|2019 \|Differentially private training in TensorFlow. \|Yes
Opacus{{Cite web\|url=https://opacus.ai/\|title=Opacus · Train PyTorch models with Differential Privacy\|website=opacus.ai\|accessdate=29 March 2023}} \|Meta \|2020 \|Differentially private training in PyTorch. \|Yes

class="sortable wikitable" style="text-align: left"

Name

! Developer

! Year Introduced

! Notes

! Still maintained?

Diffprivlib{{cite web | url=https://github.com/IBM/differential-privacy-library | title=Diffprivlib v0.5 | website=GitHub | date=17 October 2022 }}

|IBM{{Cite arXiv| eprint=1907.02444 | last1=Holohan | first1=Naoise | last2=Braghin | first2=Stefano | author3=Pól Mac Aonghusa | last4=Levacher | first4=Killian | title=Diffprivlib: The IBM Differential Privacy Library | year=2019 | class=cs.CR }}

|2019

|Python library.

|Yes

TensorFlow Privacy{{cite web |last1=Radebaugh |first1=Carey |last2=Erlingsson |first2=Ulfar |title=Introducing TensorFlow Privacy: Learning with Differential Privacy for Training Data |work=Medium |url=https://medium.com/tensorflow/introducing-tensorflow-privacy-learning-with-differential-privacy-for-training-data-b143c5e801b6 |date=March 6, 2019}}{{cite web |title=TensorFlow Privacy |website=GitHub|url=https://github.com/tensorflow/privacy|date=2019-08-09}}

|Google

|2019

|Differentially private training in TensorFlow.

|Yes

Opacus{{Cite web|url=https://opacus.ai/|title=Opacus · Train PyTorch models with Differential Privacy|website=opacus.ai|accessdate=29 March 2023}}

|Meta

|2020

|Differentially private training in PyTorch.

|Yes

Research projects and prototypes

class="sortable wikitable" style="text-align: left"
Name ! Citation ! Year Published ! Notes
PINQ: An API implemented in C#. \| {{cite journal \|last1=McSherry \|first1=Frank \|title=Privacy integrated queries \|journal=Communications of the ACM \|date=1 September 2010 \|volume=53 \|issue=9 \|pages=89–97 \|doi=10.1145/1810891.1810916 \|s2cid=52898716 \|url=https://www.microsoft.com/en-us/research/wp-content/uploads/2010/08/networking.pdf}} \| 2010 \|
Airavat: A MapReduce-based system implemented in Java hardened with SELinux-like access control. \| {{cite journal \|last1=Roy \|first1=Indrajit \|last2=Setty \|first2=Srinath T.V. \|last3=Kilzer \|first3=Ann \|last4=Shmatikov \|first4=Vitaly \|last5=Witchel \|first5=Emmett \|title=Airavat: Security and Privacy for MapReduce \|journal=Proceedings of the 7th Usenix Symposium on Networked Systems Design and Implementation (NSDI) \|date=April 2010 \|url=https://www.usenix.org/legacy/event/nsdi10/tech/full_papers/roy.pdf}} \| 2010 \|
Fuzz: Time-constant implementation in Caml Light of a domain-specific language. \| {{cite journal \|last1=Haeberlen \|first1=Andreas \|last2=Pierce \|first2=Benjamin C. \|last3=Narayan \|first3=Arjun \|title=Differential Privacy Under Fire \|journal=20th USENIX Security Symposium \|date=2011}} \| 2011 \|
GUPT: Implementation of the sample-and-aggregate framework. \| {{cite book\|last1=Mohan \|first1=Prashanth \|last2=Thakurta \|first2=Abhradeep \|last3=Shi \|first3=Elaine \|author3-link= Elaine Shi \|last4=Song \|first4=Dawn \|last5=Culler \|first5=David E. \|chapter=GUPT: Privacy Preserving Data Analysis Made Easy \|title=Proceedings of the 2012 ACM SIGMOD International Conference on Management of Data \|pages=349–360 \|doi=10.1145/2213836.2213876\|s2cid=2135755 }} \| 2012 \|
$\epsilon$ KTELO: A framework and system for answering linear counting queries. \|{{cite book \|last1=Zhang \|first1=Dan \|last2=McKenna \|first2=Ryan \|last3=Kotsogiannis \|first3=Ios \|last4=Hay \|first4=Michael \|last5=Machanavajjhala \|first5=Ashwin \|last6=Miklau \|first6=Gerome \|title=Proceedings of the 2018 International Conference on Management of Data \|chapter=EKTELO: A Framework for Defining Differentially-Private Computations \|date=June 2018 \|pages=115–130 \|doi=10.1145/3183713.3196921\|arxiv=1808.03555 \|isbn=9781450347037 \|s2cid=5033862 }} \| 2018 \|

class="sortable wikitable" style="text-align: left"

Name

! Citation

! Year Published

! Notes

PINQ: An API implemented in C#.

| {{cite journal |last1=McSherry |first1=Frank |title=Privacy integrated queries |journal=Communications of the ACM |date=1 September 2010 |volume=53 |issue=9 |pages=89–97 |doi=10.1145/1810891.1810916 |s2cid=52898716 |url=https://www.microsoft.com/en-us/research/wp-content/uploads/2010/08/networking.pdf}}

| 2010

Airavat: A MapReduce-based system implemented in Java hardened with SELinux-like access control.

| {{cite journal |last1=Roy |first1=Indrajit |last2=Setty |first2=Srinath T.V. |last3=Kilzer |first3=Ann |last4=Shmatikov |first4=Vitaly |last5=Witchel |first5=Emmett |title=Airavat: Security and Privacy for MapReduce |journal=Proceedings of the 7th Usenix Symposium on Networked Systems Design and Implementation (NSDI) |date=April 2010 |url=https://www.usenix.org/legacy/event/nsdi10/tech/full_papers/roy.pdf}}

| 2010

Fuzz: Time-constant implementation in Caml Light of a domain-specific language.

| {{cite journal |last1=Haeberlen |first1=Andreas |last2=Pierce |first2=Benjamin C. |last3=Narayan |first3=Arjun |title=Differential Privacy Under Fire |journal=20th USENIX Security Symposium |date=2011}}

| 2011

GUPT: Implementation of the sample-and-aggregate framework.

| {{cite book|last1=Mohan |first1=Prashanth |last2=Thakurta |first2=Abhradeep |last3=Shi |first3=Elaine |author3-link= Elaine Shi |last4=Song |first4=Dawn |last5=Culler |first5=David E. |chapter=GUPT: Privacy Preserving Data Analysis Made Easy |title=Proceedings of the 2012 ACM SIGMOD International Conference on Management of Data |pages=349–360 |doi=10.1145/2213836.2213876|s2cid=2135755 }}

| 2012

\epsilon

KTELO: A framework and system for answering linear counting queries.

|{{cite book |last1=Zhang |first1=Dan |last2=McKenna |first2=Ryan |last3=Kotsogiannis |first3=Ios |last4=Hay |first4=Michael |last5=Machanavajjhala |first5=Ashwin |last6=Miklau |first6=Gerome |title=Proceedings of the 2018 International Conference on Management of Data |chapter=EKTELO: A Framework for Defining Differentially-Private Computations |date=June 2018 |pages=115–130 |doi=10.1145/3183713.3196921|arxiv=1808.03555 |isbn=9781450347037 |s2cid=5033862 }}

| 2018

References

Category:Differential privacy

Category:Information privacy

List of implementations of differentially private analyses

Real-world deployments

Production software packages

= Statistical analyses =

= Machine learning =

Research projects and prototypes

See also

References