Design of a classifier consists of two stages: feature extraction and classifier learning. For a better performance, the nature, characteristics, or underlying structure of data should be taken into account in either of the stages when we design a classifier. In this thesis, we present kernel...

In recent years, fast spreading worms have become one of the major threats to the security of the Internet. Code Red, Nimda, Slammer, Blaster, MyDoom... these worms kept hitting the Internet and caused severe damage to our society. However, until now we have not fully understand the behaviors of...

In this dissertation, I develop a new theoretical framework for the modeling, pricing analysis, and computation of solutions to electric power supply chains with power generators, suppliers, transmission service providers, and the inclusion of consumer demands. In particular, I advocate the...

The problem of learning a latent model for sparse or low-dimensional representation of high-dimensional data has attracted significant attention for many years. This thesis focuses on learning latent models for sparse or low-dimensional representation of images, dynamic data, and documents with...

Prescriptive decision theory requires decision-makers to express their beliefs and tastes with explicit, precise, and complete information. This information, expressed in subjective probability and utility functions, is often difficult and costly to obtain from a decision-maker. It is obvious...

For hyperspectral data classification, the avoidance of singularity of covariance estimates or excessive near singularity estimation error due to limited training data is a key problem. This study is intended to solve problem via regularized covariance estimators and feature extraction...

This paper proposes a new approach of generation expansion planning and its risk assessment under the competitive market environment. Different from the traditional probabilistic production costing (PPC) method under monopoly environment which is based on the deterministic generator loading...

The real electricity markets are usually oligopoly, where market suppliers (generators) have some market power and can adopt strategic bidding strategies for maximum profits. Generally, the game-theory based methods are the natural way to analyze the market equilibrium and study the strategic...

We propose to apply the concept of irrelevant variability normalization to the general problem of learning structure from data. Because of the problems of a diversified training data set and/or possible acoustic mismatches between training and testing conditions, the structure learned from the...

This thesis concerns the problem of efficient resourceallocation under constraints. In many applications a finitebudget is used and allocating it efficiently can improveperformance. In the context of medical imaging the constraint is exposure to ionizing radiation, e.g., computed tomography...