PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created...
Transcript of PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created...
![Page 1: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/1.jpg)
Differential Privacy in Practice
WITSEC, Budapest, 2019
Szilvia Lestyán
CrySyS Lab, [email protected]
w w w . c r y s y s . h u
![Page 2: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/2.jpg)
|
What is privacy?
▪ Privacy is the right to private andfamily life, home and communications, to be autonomous, to be let alone
– universal human right
▪ Information privacy is the right tohave some control over how your personal information is used
2
![Page 3: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/3.jpg)
|
Why is it important?
▪ Identity fraud
▪ Your data is valuable
▪ Profiling and surveillance
▪ Stigmatization,discrimination
▪ Freedom of thought and speech
▪ Everybody has something to hide
3
![Page 4: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/4.jpg)
|
Why is it important?
▪ Identity fraud
▪ Your data is valuable
▪ Profiling and surveillance
▪ Stigmatization,discrimination
▪ Freedom of thought and speech
▪ Everybody has something to hide
4
![Page 5: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/5.jpg)
|
GDPR
▪ Personaldata:
any information related to an
identified or identifiable
natural person
▪ (re-)identification is achieved through
“identifiers”, which holds a particularly
privileged and close relationship with
the individual
5
![Page 6: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/6.jpg)
|
Identifiers
6
▪ Direct identifiers unambiguously identify a person▪ “Prime Minister of Hungary in 2017”
![Page 7: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/7.jpg)
|
Identifiers
7
▪ Direct identifiers unambiguously identify a person▪ “Prime Minister of Hungary in 2017”
▪ Indirect (quasi-)-identifier may ambiguously identify aperson– “A prime minister inEurope”
![Page 8: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/8.jpg)
|
Identifiers
8
▪ Direct identifiers unambiguously identify a person▪ “Prime Minister of Hungary in 2017”
▪ Indirect (quasi-)-identifier may ambiguously identify aperson– “A prime minister inEurope”
“A prime minister in Europe”
+
"born on May 31,1963”
![Page 9: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/9.jpg)
|
Some identifiers
9
▪ GDPR referstoall personaldataas identifierswhichtogether unambiguously identify a person in the givencontext
![Page 10: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/10.jpg)
|
Identifiable
▪ A person is identifiable:
“To determine whether a natural person is identifiable, account should be taken of all the means reasonably likely to be used, such as singling out, either by the controller or by another person to identify the natural person directly or indirectly. “
10
![Page 11: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/11.jpg)
|
What does itmean?
11
▪ A person is identifiable, if:
– Plausible attack: The attacker has enough motivation to launch theattack…
– Reasonable chance of succeeding: the success probability of theattack is high enough
▪ There are NO explicit pre-defined thresholds of plausibility and reasonable chance in GDPR, as it is context-dependent
![Page 12: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/12.jpg)
|
Example
13
▪ A hospital in Michigan publishes a medical dataset with 3
attributes: (1) ZIP, (2) Age, (3) Sex, (4)Diagnosis
▪ It is personal data according tothe GDPR
▪ Microdata (individuals are identified)
Removing
names?
![Page 13: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/13.jpg)
|
Example
14
▪ Re-identification attack:1. Purchase the voter registration data for $10
Zipcode Age Sex Disease
47677 29 F Ovarian Cancer
47602 22 F Ovarian Cancer
47678 27 M Prostate Cancer
47905 43 M Flu
47909 52 F Heart Disease
47906 47 M Heart Disease
Name Zipcode Age Sex
Alice 47677 29 F
Bob 47983 65 M
Carol 47677 22 F
Dan 47532 23 M
Ellen 46789 43 F
Microdata Voter registration data
![Page 14: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/14.jpg)
|
Example
15
▪ Re-identification attack:1. Purchase the voter registration data for $10
2. Associate a voter record with the corresponding medical record (along with matching ZIP, Age, Sexattributes)
Zipcode Age Sex Disease
47677 29 F Ovarian Cancer
47602 22 F Ovarian Cancer
47678 27 M Prostate Cancer
47905 43 M Flu
47909 52 F Heart Disease
47906 47 M Heart Disease
Name Zipcode Age Sex
Alice 47677 29 F
Bob 47983 65 M
Carol 47677 22 F
Dan 47532 23 M
Ellen 46789 43 F
Microdata Voter registration data
![Page 15: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/15.jpg)
|
Example
16
▪ Re-identification attack:1. Purchase the voter registration data for $10
2. Associate a voter record with the corresponding medical record (along with matching ZIP, Age, Sexattributes)
▪ Success probability of linking: 63%*
Zipcode Age Sex Disease
47677 29 F Ovarian Cancer
47602 22 F Ovarian Cancer
47678 27 M Prostate Cancer
47905 43 M Flu
47909 52 F Heart Disease
47906 47 M Heart Disease
Name Zipcode Age Sex
Alice 47677 29 F
Bob 47983 65 M
Carol 47677 22 F
Dan 47532 23 M
Ellen 46789 43 F
Microdata Voter registration data
![Page 16: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/16.jpg)
|
Example II. – Query Auditing
17| 17
▪ Given a database with some disclosure policy
– HIV is private, but aggregated values of HIV records may beavailable
– e.g. SUM, COUNT, MEDIAN, MAX
Name Sex ZIP Blood sugar HIV
John S. Male 1123 4.3 True
John D. Male 1123 5.2 False
Jerry K. Male 1114 6.1 True
Jack. D. Male 8423 3.2 False
Eve A. Female 1234 7.1 True
![Page 17: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/17.jpg)
|
Example II. – Query Auditing
18
solve a system of linear equations
|
▪ SUM(HIV) WHERE ZIP < 8000– x1 + x2 + x3 + x4 = 2
▪ SUM(HIV) WHERE ZIP = 1123– x1 + x2 =1
▪ SUM(HIV) WHERE ZIP > 1200– x4 + x5 =1
x1
3
1 1 1 1 0
1 1 0 0 0
0 0 0 1 1 x 4
x5
x2
x =
2
1
1
where xi in {0, 1} for all i
![Page 18: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/18.jpg)
|
Example II. – Query Auditing
19
solve a system of linear equations
|
▪ SUM(HIV) WHERE ZIP < 8000– x1 + x2 + x3 + x4 = 2
▪ SUM(HIV) WHERE ZIP = 1123– x1 + x2 =1
▪ SUM(HIV) WHERE ZIP > 1200– x4 + x5 =1
x1
3
1 1 1 1 0
1 1 0 0 0
0 0 0 1 1 x 4
x5
x2
x =
2
1
1
where xi in {0, 1} for all i
MATHS
MAGIC
![Page 19: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/19.jpg)
|
Example II. – Query Auditing
20
solve a system of linear equations
|
▪ SUM(HIV) WHERE ZIP < 8000– x1 + x2 + x3 + x4 = 2
▪ SUM(HIV) WHERE ZIP = 1123– x1 + x2 =1
▪ SUM(HIV) WHERE ZIP > 1200– x4 + x5 =1
x1
3
1 1 1 1 0
1 1 0 0 0
0 0 0 1 1 x 4
x5
x2
x =
2
1
1
where xi in {0, 1} for all i
MATHS
MAGICPRIVACY
BREACH
![Page 20: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/20.jpg)
|
Philosophy of Differential privacy
21
▪ Absolute (Perfect) privacy:access to the published data should
not enable the adversary to learn anything extra about any
individual compared to no access to the data
![Page 21: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/21.jpg)
|
Philosophy of Differential privacy
22
▪ Absolute (Perfect) privacy:access to the published data should
not enable the adversary to learn anything extra about any
individual compared to no access to the data
There is always some background knowledge which allows
absolute privacy breach
This is
unachievable in
practice!
![Page 22: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/22.jpg)
|
Philosophy of Differential privacy
23
▪ AVG = 175 cm
▪ John is +10
Privacy breach?
![Page 23: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/23.jpg)
|
PRIVACY BREACH!
(in absolute sense)
• John has cancer with 50% and smokes
• Study shows smoking causes cancer in 92%
24
![Page 24: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/24.jpg)
|
BUT!
25
Does removing John from the database cause any change in the outcome?
![Page 25: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/25.jpg)
|
Philosophy Differential privacy
▪ Theoutcome of theanonymization
scheme should be more or less
independent of thevalueof a single record
➢ Itdoes not “leak” informationabout any
single record
➢ It can formally beproven!
▪ Differentialprivacyaims tohide only those
information that is specifictoJohn (orany
single individual in thedataset)
26
![Page 26: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/26.jpg)
|
Differential privacy
Tell me f(x)
f(x)+noisex1…xn
![Page 27: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/27.jpg)
|
GDPR vs.
Differential Privacy
• DP implies GDPR (not vicaversa)
• Identifiers vs noise• In GDPR DP is a method• The confidence of ALL inferences
are bounded
• Formally provable!
• Apple, Uber, Google
• Aggregates vs Microdata?
• ”Syntactic”guarantees are • Not sufficient inpractice• Does not defend from future
attacks• Not formally provable
![Page 28: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/28.jpg)
|
Conclusions
29
So what to use?
![Page 29: PowerPoint-bemutatóTitle: PowerPoint-bemutató Author: ��Lesty�n Szilvia Created Date: 10/10/2019 11:19:57 PM](https://reader033.fdocuments.us/reader033/viewer/2022051923/60116bd7fc95e83300576d52/html5/thumbnails/29.jpg)
|
ConclusionsSo what to use?
Experts!
30