Differential Privacy - Revision history

Tom: /* Solutions */

2022-04-25T23:28:38Z

‎Solutions

Tom: /* Solutions */

2022-04-25T23:22:33Z

‎Solutions

Tom: /* Federation Learning and Analysis */

2022-04-13T22:54:35Z

‎Federation Learning and Analysis

Tom: /* Solutions */

2022-04-13T22:52:12Z

‎Solutions

Tom: /* Problems */

2021-01-12T18:24:26Z

‎Problems

Tom: /* Solutions */

2021-01-12T18:18:38Z

‎Solutions

Tom: /* Context */

2020-03-25T17:54:06Z

‎Context

Tom: /* Full Title or Meme */

2020-03-25T17:52:10Z

‎Full Title or Meme

Tom: /* Full Title or Meme */

2019-11-11T19:28:05Z

‎Full Title or Meme

Tom: /* References */

2019-11-11T19:27:38Z

‎References

@@ Line 21: / Line 21: @@
 In September 2019, Google released an open source version of the differential privacy library, making the capability generally available. To date, differential privacy has been adopted by the US Census Bureau, along with a number of technology companies.
-[[Differential Privacy]] is basically just a technique for adding noise to a signal. Shannon showed how much noise can be removed by redundancy which tells us that the more noise added, the less the signal that survives. "In the context of differentially private data analysis there is a trade-off between privacy and utility.  In the context of differentially private cryptographic primitive and resulting applications, there is a broader trade-off space between privacy, utility and performance."<ref>Sameer Wagh +3, ''DP-Cryptography: Marrying Differential Privacy and Cryptography in Emerging Applications'' (2021-02) '''CACM 64''' No 2 pp 84ff</ref>
+[[Differential Privacy]] is basically just a technique for adding noise to a signal. Shannon showed how much noise can be removed by redundancy which tells us that the more noise added, the less the signal that survives. "In the context of differentially private data analysis there is a trade-off between privacy and utility.  In the context of differentially private cryptographic primitive and resulting applications, there is a broader trade-off space between privacy, utility and performance."<ref>Sameer Wagh +3, ''DP-Cryptography: Marrying Differential Privacy and Cryptography in Emerging Applications'' (2021-02) '''CACM 64''' No 2 pp 84ff</ref> The same issue of CACM contains new algorithms designed to recover just such noisy data in data bases.<ref>Abolfazi Asudeh +6, ''Scalable Signal Reconstruction for a Broad Range of Applications''  (2021-02) '''CACM 64''' N0 2 pp106 ff</ref>
 ===Federation Learning and Analysis===
 An exploration of how data ownership and provenance can be made first-class concepts in systems for learning and analytics in areas know as Federation Learning and Analysis.<ref>Kallista Bonawitz +3, ''Federation Learning and Privacy''. '''CACM 65''' no 5 (2022-04) p 90ff</ref> This allows multiple devices, some in user's hands, to collaborate in creating a common knowledge base.  Consider the ability of smart phone to complete sentences. This has the potential to leak data unless carefully designed.

@@ Line 20: / Line 20: @@
 In September 2019, Google released an open source version of the differential privacy library, making the capability generally available. To date, differential privacy has been adopted by the US Census Bureau, along with a number of technology companies.
 ===Federation Learning and Analysis===
-An exploration of how data ownership and provenance can be made first-class concepts in systems for learning and analytics in ares know as Federation Learning and Analysis.<ref>Kallista Bonawitz +3, ''Federation Learning and Privacy''. '''CACM 65''' no 5 (2022-04) p 90ff</ref> This allows multiple devices, some in user's hands, to collaborate in creating a common knowledge base.  Consider the ability of smart phone to complete sentences. This has the potential to leak data unless carefully designed.
+An exploration of how data ownership and provenance can be made first-class concepts in systems for learning and analytics in areas know as Federation Learning and Analysis.<ref>Kallista Bonawitz +3, ''Federation Learning and Privacy''. '''CACM 65''' no 5 (2022-04) p 90ff</ref> This allows multiple devices, some in user's hands, to collaborate in creating a common knowledge base.  Consider the ability of smart phone to complete sentences. This has the potential to leak data unless carefully designed.
 ==References==

@@ Line 21: / Line 21: @@
 In September 2019, Google released an open source version of the differential privacy library, making the capability generally available. To date, differential privacy has been adopted by the US Census Bureau, along with a number of technology companies.
 ===Federation Learning and Analysis===
-An exploration of how data ownership and provenance can be made first-class concepts in systems for learning and analytics in ares know as Federation Learning and Analysis.<ef>Kallista Bonawitz +3, ''Federation Learning and Privacy''. '''CACM 65''' no 5 (2022-04) p 90ff</ref>
+An exploration of how data ownership and provenance can be made first-class concepts in systems for learning and analytics in ares know as Federation Learning and Analysis.<ref>Kallista Bonawitz +3, ''Federation Learning and Privacy''. '''CACM 65''' no 5 (2022-04) p 90ff</ref> This allows multiple devices, some in user's hands, to collaborate in creating a common knowledge base.  Consider the ability of smart phone to complete sentences. This has the potential to leak data unless carefully designed.
 ==References==

← Older revision		Revision as of 22:52, 13 April 2022
Line 20:		Line 20:

	In September 2019, Google released an open source version of the differential privacy library, making the capability generally available. To date, differential privacy has been adopted by the US Census Bureau, along with a number of technology companies.		In September 2019, Google released an open source version of the differential privacy library, making the capability generally available. To date, differential privacy has been adopted by the US Census Bureau, along with a number of technology companies.
		+	===Federation Learning and Analysis===
		+	An exploration of how data ownership and provenance can be made first-class concepts in systems for learning and analytics in ares know as Federation Learning and Analysis.<ef>Kallista Bonawitz +3, ''Federation Learning and Privacy''. '''CACM 65''' no 5 (2022-04) p 90ff</ref>

	==References==		==References==

@@ Line 6: / Line 6: @@
 ==Problems==
-To much noise can render the data unless. One analysis showed that a differentially private verson of the 2010 census included households that supposedly had 90 people.<ref>Angel Chen, ''Differential Privacy.'' (2020-03) Technology Review p 27.</ref>
+To much noise can render the data unless. One analysis showed that a differentially private verson of the 2010 census included households that supposedly had 90 people.<ref>Angel Chen, ''Differential Privacy.'' (2020-03) Technology Review p 27.</ref> It seems that "normal statistics" biases in the randomization would hide any data that was not "normal".
 ==Solutions==

@@ Line 3: / Line 3: @@
 ==Context==
-[[Differential Privacy]] is a mathematical technique that injects inaccuracies or "noise," into the data, making some people younger and an equal number older, changing races or other attributes. The more noise you inject, the harder deanonymization becomes. Apple and Facebook started using this technique in 2019 to collect aggregate data without identifier particular users.
+[[Differential Privacy]] is a mathematical technique that injects inaccuracies or "noise," into the data, making some people younger and an equal number older, changing races or other attributes. The more noise you inject, the harder deanonymization becomes. Apple and Facebook started using this technique in 2019 to collect aggregate data without identifing particular users.
 ==Problems==

@@ Line 1: / Line 1: @@
 ==Full Title or Meme==
 [[Differential Privacy]] is a system for publicly sharing information about a data collection by describing the patterns of groups within the collection while withholding information that could identify the [[Subject]].
 ==Solutions==

@@ Line 2: / Line 2: @@
 [[Differential privacy]] is a system for publicly sharing information about a data collection by describing the patterns of groups within the collection while withholding information that could identify the [[Subject]].
 ==Solutions==
 ==References==
-U.S. Census Bureau officials said the agency is revamping its systems to prevent anyone from using published data to target individual respondents through the information they disclosed to the census. The bureau aims to use a mathematical process, called [[Differential Privacy]], to modify census results sufficiently to reliably conceal respondents' identity. The agency will make small additions to and subtractions from each number, prior to almost every table's publication, and significantly cut the number of published statistics. Although data users are concerned these changes will disrupt their use of census data, not addressing the danger could allow information on individuals to be exposed, violating federal privacy law and elevating the risk of identity theft and other kinds of misuse.
 [[Category:Glossary]]
 [[Category:Privacy]]