Computational Algorithms and Theories for Learning with Big Data
The scale and dimensionality of data associated with machine learning applications have seen unprecedented growth in the last decade. Although it has brought great opportunities in various domains, there still remain many challenges for solving big data learning problems. In this presentation, I will focus on the computational perspective for learning with big data, and will talk about challenges and opportunities for solving optimization problems arising in machine learning. In particular, I will present a new stochastic optimization algorithm for solving large-scale problems, and randomized machine learning algorithms for tackling high-dimensional data. I will also discuss opportunities for learning with big data through the lens of computational learning theory.