[Colloquium] Research at TTIC: Anand Sarwate

[Liv Leader]

When:     Friday, April 20 @ 12

Where:    TTIC, 6045 S. Kenwood Avenue, Room #526

Who:      Anand Sarwate

Title:       Differentially private machine learning

The exponential increase of personal data stored in electronic databases
has raised important questions about privacy from both a policy and
technological perspective.  Many approaches to enforcing privacy in data
publishing are susceptible to attacks that allow a malicious member of the
public to compromise and re-identify individuals.  Differential privacy is
a way of quantifying privacy risks incurred while computing using private
data that is resistant to these attacks.  In this talk I will discuss some
recent work on learning using private data while guaranteeing differential
privacy.  Ensuring privacy entails losing some accuracy, and we quantify
the results and validate them on data sets.  I will discuss approaches to
learning classifiers via empirical risk minimization (ERM), including
support vector machines (SVMs), principle components analysis (PCA), and
possibly applications to synthetic data generation.

Joint work with Kamalika Chaudhuri, Claire Monteleoni, and Staal Vinterbo

-- 
Liv Leader
Human Resources Coordinator

Toyota Technological Institute Chicago
6045 S Kenwood Ave
Chicago, IL 60637
Phone- (773) 702-5033
Fax-     (773) 834-9881
Email-  lleader at ttic.edu <jam at ttic.edu>
Web-   www.ttic.edu
<http://www.ttic.edu/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20120417/232112ae/attachment.htm