Time: 25/1/2014 (9:00-12:00) and 27/1/ 2014 (10:00 – 12:00)

Location: VIASM Lecture Hall (C2).

Lecturer: Prof. Vu Ha Van, Yale University

Abstract:

A common modeling assumption in many engineering applications is that the underlying data lies (approximately) on a low-dimensional linear subspace. Furthermore, real-life data is often corrupted with large errors or can even be incomplete. (One can think of a blurred image from a satellite, for instance.)

A fundamental question is to recover the real data from noisy, incomplete version we observe. A strongly connected problem is to compress data, which means instead of keeping a huge number of bits, we only need to keep a small portion of them, but will be able to recover the rest if necessary. This can lead to fundamental progresses in medical science, among others, a considerable speeding up of MRI techniques.

As data is usually represented by a large matrix, certain mathematical tools from linear algebra, probability, and optimization become crucial. In this seminar series, I am going to introduce the most basic ones and their applications. Among real-life applications, I am going to discuss

(1) The Page rank algorithm: The Pagerank algorithm, developed by Larry Page and Sergey Brin, is the core of Google’ search engine, perhaps the most useful software on earth. This algorithm enables one a rank webpages quickly and correctly. We are going to cover this algorithm in details, which requires no more than college level knowledge of linear algebra and probability.

(2) Candes-Tao’s work on compress sensing and potential applications in MRI industry. How to make an one pixle camera.

(3) The 1.000.000 dollars Netflix problem.

Registration:

Please fill in the registration form below your contact information.