Abstract:
Principal component analysis (PCA) is often used to reduce the dimension of data by selecting a few orthonormal vectors that explain most of the variance structure of the data. In order to build a robust model, L1 PCA uses the L1 norm to measure the objective, whereas the conventional PCA uses the L2 norm. For two types of L1 PCA problems, maximization of the dispersion of the projected data and minimization of the fitting error of the reconstructed data, we provide iterative algorithms and analyses, and compare their performances against benchmark algorithms in the literature.
Bio:
Diego Klabjan is a professor at Northwestern University, Department of Industrial Engineering and Management Sciences. He is also Founding Director, Master of Science in Analytics. He obtaining his doctorate from the School of Industrial and Systems Engineering of the Georgia Institute of Technology which is consistently the top ranked department. His research is focused on optimization and machine learning with concentration in transportation, healthcare, and web. Professor Klabjan has led projects with large companies such as FedEx Express, BNSF, General Motors, United Continental, and many others, and he is also assisting numerous start-ups with their analytics needs. He is also a founder of Opex Analytics LLC.