Workshop on Mathematics for Defence
(13 Apr 2012)

Jointly organized with The Defence Research & Technology Office (DRTech)

~ Abstracts ~


Mathematical methods for image/video restoration
Ji Hui, National University of Singapore

In recent years, there have been great processes on the development of innovative mathematical models and new mathematical methods applied to many challenging problems arising in image processing and computer vision. In this talk, I will discuss the theory and numerical methods of sparse approximation under tight frame and its applications in image/video restoration.

« Back...


Symmetric-key cryptography
Thomas Peyrin, Nanyang Technological University

In this talk will be described from a high level view current state-of-the-art designs of two of the most important symmetric key cryptography primitives: block ciphers and hash functions. These bricks are the backbones of the security applications in various industries such as telecommunications, banking, access control, etc. and used in everyday applications such as PC, cellphones, smart-cards, Internet, etc. PC, cellphones, smart-cards, Internet, etc. A novel and promising type of diffusion matrices will also be covered, which allow to create secure block ciphers and hash functions with reduced hardware footprint.

« Back...


Mathematical theory and numerical methods for the study of rare events
Ren Weiqing, National University of Singapore and Institute of High Performance Computing, A*STAR

Many processes from applied sciences are driven by rare but important events. Well known examples include nucleation events during phase transitions, conformational changes of bio-molecules, chemical reactions, etc. The system is confined in metastable states for long times before making transitions from one metastable state to another. The presence of disparate time scales makes the study of transition pathways and transition rates a very challenging task. In this talk, I will discuss mathematical theory and numerical methods for the study of complex energy landscapes and the associated barrier-crossing events.

« Back...


An overview of variable selection in high dimensional data analysis
Tang Chengyong, National University of Singapore

Various high dimensional data are increasingly available along with drastically technological development. In traditional statistical analysis, we assume many observations and few variables. Nowadays the trend is that we have more observations, but even more variables are collected. This talk overviews some variable selection methods for high dimensional data analysis. The merits of applying the variable selection method will be demonstrated through some examples.

« Back...

Best viewed with IE 7 and above