[Busec] Course announcement at Boston University

Mayank Varia varia at bu.edu
Tue Jan 5 10:55:46 EST 2016


Two courses covering topics in security, cryptography, and privacy will be
given at Boston University in the spring semester: Symmetric cryptography:
design & practice and Privacy Tools for Data Science. The courses will
present very different problems and solutions, and students are encouraged
to exploit this opportunity to experience a variety of aspects of the
security and privacy space. See details below.


CS591 (Computer Science Topics) Symmetric cryptography: design & practice.
Instructor: Mayank Varia.
Tuesday, Thursday 2:00-3:30PM. PSY B53.

Description: This course will introduce the techniques in the theory,
design, and cryptanalysis of symmetric cryptography primitives. We will
examine several primitives including stream ciphers, block ciphers, and
collision-resistant hash functions; specific ciphers studied in detail
include DES, AES, and the SHA family of hash functions. Additionally, we
will analyze the mathematical strength of these primitives toward common
types of mathematical cryptanalysis. Finally, we will explore
provably-secure constructions of symmetric-key encryption schemes and
message authentication codes from these building blocks. The course will
have a hands-on approach, culminating with a symmetric cipher competition
of our own. The most important prerequisite is familiarity with algebra and
probability; prior exposure to cryptography is helpful but not required.


CS591 (Computer Science Topics) Privacy Tools for Data Science.
Instructors: Kobbi Nissim and George Kollios.
Monday, 2:30-5:30PM. PSY B43.

Description: Our world is data driven. Information about us is constantly
collected and analyzed for research purposes, to provide us ads and
services, to inform policy, etc. These uses of data raise concerns about
individual privacy. We will explore data privacy in the context of data
analytics landscape, focusing on a collection of available and future
technological solutions.

 In particular, we will explore some points of the current technological
privacy landscape:
 - What is data privacy?
 - Data anonymization techniques and their vulnerabilities.
 - Differential Privacy.
 - Basic data science techniques: classification, clustering, and
prediction.
 - Secure computation and data mining
 - Differential Privacy and data analytics
 - Advanced topics (time permitting)

The target audience audience for the course are students interested in
performing machine learning tasks on collections of sensitive individual
data.  Needed background includes knowledge of algorithms, linear algebra,
and probability at the level of undergraduate studies. Prior knowledge in
machine learning/privacy/cryptography will be helpful but not necessary.
Students will be required to hand in 3-4 homework sets and present a final
project.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs-mailman.bu.edu/pipermail/busec/attachments/20160105/748e6008/attachment.html>


More information about the Busec mailing list