# |
Date |
Topic |
Readings |
MODULE 1: Introduction |
|||
1 |
Tue, Aug 30 |
Overview on Privacy & Learning (slides) |
"A Face is exposed for AOL Searcher No. 4417749", New York Times, Aug 2006 "How Companies Learn your Secrets", New York Times, Feb 2012 |
2 |
Thu, Sep 1 |
Intro to Differential Privacy (slides) |
"Calibrating Noise to Sensitivity in Private Date Analysis" C. Dwork, F. McSherry, K. Nissim, A.Smith TCC 2006 Optional: "Differential Privacy" C. Dwork, ICALP 2006 |
Tue, Sep 6 |
NO CLASS |
||
Thu, Sep 8 |
NO CLASS |
||
3 |
Tue, Sep 13 |
What is (and is not) privacy? (slides) |
|
4 |
Thu, Sep 15 |
Privacy and composition (slides) |
"Revealing information while
preserving privacy" I. Dinur, K. Nissin PODS 2003 Optional: "Composition Attacks and Auxiliary Information in Data Privacy" S. Ganta, S. Kasiviswanathan, A. Smith KDD 2008 |
5 |
Tue, Sep 20 |
Differential Privacy and Privacy (slides) |
"No Free Lunch in Data Privacy", D. Kifer, A. Machanavajjhala SIGMOD 2011 |
MODULE 2: DP Algorithms |
|||
6 |
Thu, Sep 22 |
Building Blocks (slides) |
"The algorithmic foundations of Differential Privacy", Roth, Dwork (sections 3.3, 3.4, 3.5) |
7 |
Tue, Sep 27 |
Smooth Sensitivity and Sample Aggregate Framework (slides) |
"Smooth Sensitivity and Sampling in Private Data Analysis", K. Nissim, S. Raskhodnikova, A. Smith STOC 2007 |
8 |
Thu , Sep 29 |
Answering sets of queries (slides) (Deadline for choosing project topics) |
"Boosting the Accuracy of Differentially Private Histograms Through Consistency", M. Hay, V. Rastogi, G. Miklau, D. Suciu, PVLDB 2010 "Differential Privacy via Wavelet Transforms", X. Xiao, G. Wang, J. Gehrke ICDE 2009 |
9 |
Tue, Oct 4 |
Matrix Mechanism and Data dependent algorithms (slides) |
|
10 |
Thu Oct 6 |
Multiplicative Weights and Differential Privacy (slides) | "A simple and practical algorithm for differentially private data release", F. McSherry, K. Ligett, M. Hardt, NIPS 2012 |
Tue Oct 11 |
NO CLASS: FALL BREAK |
||
11 |
Thu, Oct 13 |
Sparse Vector Technique and online query answering (slides) (Deadline for submitting project proposal) |
"The algorithmic foundations of Differential Privacy", Roth, Dwork (sections 3.6, 4.2) Optional: "A Multiplicative Weights Mechanism for Privacy Preserving Data Analysis" M. Hardt, G. Rothblum, FOCS 2010 |
12 |
Tue Oct 18 | Customizing Privacy to applications | "Pufferfish: A Framework for Mathematical Privacy Definitions", D. Kifer, A. Machanavajjhala TODS 2013 (sections 7,8,10 optional) |
MODULE 3: Real world use cases and applications |
|||
13 |
Thu Oct 20 |
Privately collecting data from individuals |
"RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response", Erlingsson, Pihur, Korolova, ACM CCS 2014 |
14 |
Tue Oct 25 |
Releasing Census Data |
"Privacy: Theory meets practice on the map", A. Machanavajjhala, D. Kifer, J. Abowd, J. Gehrke, L. Vilhuber, ICDE 2008 "Formal privacy protection for data products combining individual and employer frames" Haney et al |
15 |
Thu, Oct 27 |
Location Privacy |
|
16 |
Tue, Nov 1 |
Privacy in graphs |
"Private Analysis of Graph Structure", Karwa et al VLDB 2011 "Publishing Graph Degree Distribution with Node Differential Privacy", Wei-Yen Day et al, SIGMOD 2016 |
17 |
Thu, Nov 3 | Implementing Differential Privacy |
"Differential Privacy Under Fire". Haeberlen, Pierce, and Narayan, 2011 "On the significance of least significant bits for differential privacy", Mironov CCS 2012 |
|
MODULE 4: DP & Learning |
||
18 |
Tue, Nov 8 |
Model Inversion: Privacy attacks on Learning (Deadline for submitting mid-term report) |
"Privacy in Pharmacogenetics: An end-to-end use case study of personalized warfarin dosing", Fredrickson et al USENIX Security 2014 |
19 |
Thu Nov 10 | Classification and Regression under Differential Privacy | "Differentially Private Empirical Risk Minimization", K. Chaudhuri, C. Monteleoni, A. Sarwate JMLR 2011 |
20 |
Tue, Nov 15 |
DP + Deep learning | "Deep learning with Differential Privacy", Abadi et al |
21 |
Thu, Nov 17 |
What can you learn privately | "What can we learn privately", Kasiviswanathan et al, FOCS 2008 |
22 |
Tue, Nov 22 |
Adaptive Analysis | "Preserving Statistical Validity in Adaptive Data Analysis", Dwork et al, STOC 2015 |
Thu, Nov 24 |
NO CLASS: THANKSGIVING |
||
23 |
Tue, Nov 29 |
Stability and generalization | |
Thu, Dec 1 |
FINAL PRESENTATIONS |