[CCICADA-announce] Seminar - November 17, 2015 - Guang Yang, IIS, Tsinghua University

[Linda Casals]

******************************************************************
     Department of Management Science and Information Systems
                    Rutgers Business School
            Rutgers, The State University of New Jersey
                      ** Department Seminar **  
                 http://msis-rutgers.net/seminar/
******************************************************************

Title: Streaming Computation: Algorithms and Lower Bounds

Speaker: Guang Yang, IIS, Tsinghua University

Date/Time: Tuesday, November 17, 2015,  1:30pm

Location: Livingston Campus at  BRR-5038
 (and also broadcasted at 1WP-302)

Host: Periklis Papakonstantinou
  
*Please check link for updates: http://msis-rutgers.net/seminar/#talk-20151017 

******************************************************************

Abstract:

Streaming algorithms are able to handle Big Data and hence they have
attracted broad research interests in the past several years. Formally,
streaming algorithms treat the massively long input as a stream that
provides only sequential access and aim to compute functions with a
small local random-access working memory and very few passes over
streams. There are various streaming settings using Read-Only or
Read-Write streams and altering the following parameters: p total number
of passes, s working memory size, t number of RW streams. Typically, p
and t are constants, while s is also very small, e.g. polylog(n) or
n^{0.1}. It is of great practical and theoretical interest to study the
ability and limitations of different streaming settings.

In this talk, I will present algorithms and impossibility results
starting from the simplest single-stream setting to the involved
multi-(RW)-stream setting. The talk is a tutorial introducing the basics
of streaming computation with a slight bias towards topics I find most
exciting.
 
About the Speaker:

Guang Yang is a graduating PhD student. He is reading computer science
and mathematics at Andrew Yao's institute at Tsinghua University under
the supervision of Periklis Papakonstantinou. His research interests are
in the foundations of computer science and is intrigued by problems with
immense practical motivation. Before he becomes a PhD student he did a
Bachelors in Engineering at the Tsinghua's Elite CS class, aka Yao
Class. He has received multiple distinctions as a PhD, undergraduate,
and high-school student including a gold medal in CMO (China Mathematics
Olympiad) for which he was offered direct admission the the
undergraduate program in Tsinghua.