[Busec] Fwd: BU course on adaptive data analysis

Ran Canetti canetti at bu.edu
Thu Sep 7 14:21:41 EDT 2017

Just In case you managed not to know of this course up till now...

-------- Forwarded Message --------
Subject: 	BU course on adaptive data analysis
Date: 	Mon, 4 Sep 2017 23:06:09 -0400
From: 	Adam Smith <ads22 at bu.edu>
To: 	Dwork, Cynthia <dwork at seas.harvard.edu>, Salil Vadhan 
<salil-privacytools at g.harvard.edu>, Jon Ullman <jullman at gmail.com>, 
Leonid Reyzin <reyzin at cs.bu.edu>, Ran Canetti <canetti at bu.edu>, Sharon 
Goldberg <goldbe at cs.bu.edu>, Vinod Vaikuntanathan 
<vinodv at csail.mit.edu>, Constantinos Daskalakis <costis at csail.mit.edu>, 
Kolaczyk, Eric D <kolaczyk at bu.edu>, Nazer, Bobak <bobak at bu.edu>, 
pi at bu.edu, cuhler at mit.edu, Madhu Sudan <madhusudan at g.harvard.edu>, 
Shelat, Abhi <a.shelat at northeastern.edu>, orecchia at bu.edu, 
Raskhodnikova, Sofya <sofya at bu.edu>, aene at bu.edu, b at boazbarak.org, 
rakhlin at wharton.upenn.edu, rigollet at math.mit.edu

Dear colleagues,

I will be teaching a graduate course at BU this semester on adaptive
data analysis and associated topics, including differential privacy.
Aaron Roth (at Penn) and I are developing the materials together.

Please circulate to interested students or relevant lists! Auditors
from other universities are welcome.

Best regards,


CS 591: Adaptive Data Analysis, Algorithmic Stability and Privacy

Times: Mondays and Wednesdays, 4:30-5:45pm
            (Lectures start this Wednesday, September 6.)

Location: MCS Building B25, Boston University
             (111 Cummington Mall, Boston)

Description: Adaptive data analysis is the increasingly common
practice by which insights gathered from data are used to inform
further analysis of the same data sets. This is common practice both
in machine learning, and in scientific research, in which data-sets
are shared and re-used across multiple studies. Standard theory
assumes that the analysis to be performed on a data set is selected
independently of the data set. Unfortunately, when the set of analyses
run is itself a function of the data, much of this theory becomes
invalid. The resulting disconnect has been blamed as one of the causes
of the crisis of reproducibility in empirical science.

This course will look at recently developed approaches to this
problem. We will see approaches stemming from the literature on
"differential privacy", approaches rooted in measuring leaked
information, and approaches coming from more standard statistical

The course is aimed at graduate students in computer science,
statistics and electrical engineering. The prerequisites are a solid
background in probability, and general "mathematical maturity"
(comfort reading and writing definitions, theorems and proofs). The
course will involve reading and reviewing research papers, pencil and
paper assignments, and some programming problems.

Aaron Roth at U. Penn and I are developing the materials (lecture
notes, homework) together, but each delivering lectures locally.
Materials will be posted here:

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs-mailman.bu.edu/pipermail/busec/attachments/20170907/394efe8d/attachment.html>

More information about the Busec mailing list