Statistical learning and Computational statistics
Statistical learning and machine learning combine theory and practice in statistics, informatics and optimization. The methods learn from observed data, find the underlying patterns, predict and solve problems with prediction uncertainty. High-dimensional integration constitutes the core of many modern estimation problems. Effective computational algorithms make it possible to apply more realistic and interesting statistical models.
Main content
Computer science is merged in statistical learning with focus on data acquisition, retrieval, mining, and reporting. Various software, including R, Phyton and Matlab are practical tools in statistical learning. The learning problems can mainly be categorized as either "supervised" or "unsupervised". Aim of supervised learning is to predict the value of an output variable based on a series of input variables, and to minimize prediction errors. The goal of unsupervised learning is to describe associations and patterns among a set of variables. Statistical machine learning includes regression, classification, density estimation, main component analysis, sparse learning, neural networks, boosting and Bayesian inference.
Members in Statistics groups are active in researches of methodology modification and in statistical learning. Research topics including sparse Bayesian learning, boosting and effective computational algorithms .
We also offer course in Statistical Learning. More information of the course, please contact Yushu Li.
