Pattern recognition: feature engineering and (deep) feature learning

Feature extraction is one of the most important steps in any pattern recognition tasks. The traditional approach is to design different types of local and global function to build-up the feature map. In contrast, the new deep architecture of convolutional neural networks automatically forms the feature maps by learning convolution operators. How the two approaches are similar and different is one main topic of this tutorial. The second topic of the talk is how context information is used in recognition tasks. Example in optical character recognition will be used to characterize the difference between the traditional dictionary/language model and recently emerging long sort-term memory networks.

Duc-Dung NGUYEN received the Bachelors degree in mathematics in 1994. He received the Masters and Ph.D. degrees in knowledge science from the Japan Advanced Institute of Science and Technology, Japan, in 2003 and 2006, respectively.
He was a Research Engineer at KDDI R&D Laboratories Inc., Japan, from 2007 to 2009. He is now with the Institute of Information Technology, Vietnam Academy of Science and Technology, Ha Noi, Vietnam. His research interests include machine learning, pattern recognition, and data mining. Dr. Nguyen was awarded the Innovative Medal from the Youth Union of Vietnam in 1998 for developing the first Vietnamese optical character recognition software, and the Technical Support Achievement Award in 2008 for his contributions at KDDI R&D Laboratories.