The Mini-course “High dimensional geometry and Data analysis” by Prof. Vu Ha Van (Yale University) has opened on July 14, 2014. This is one more event of research group Discrete Mathematics led by Assoc. Prof. Phan Thi Ha Duong.

There have been more than 30 people attending the course, which lasts until the morning of July 18, 2014. Big Data has been one of the main topics of science in the last ten years, and will be in the next several decades.

Data is usually given in a (huge) matrix form. (For instance, a 1GB digital picture is a matrix with one billion entries). The problem is to process and obtain information from these huge and often noisy matrices. Quickly and accurately! This leads to a vast collection of new mathematical challenges.

In these lectures, Prof. Van aims to cover some of the key mathematical tools that appear very useful in data analysis. These include both classical and new results in high dimensional geometry, random matrix theory, functional analysis, and probability. Most of these results are of independent interest, and in fact have been studied for a long time for entirely different, more theoretical, purposes.

Participants of the course also discuss several applications, the main one will be the community detection problem. How to partition a large group of people into smaller communities such that people in the same community behave roughly the same way.

Ten two-hour lectures will be given every five days until July 18, 2014. This is also my personal answer to the popular question: How useful is modern mathematics?