Machine Learning

I am interested in techniques for computational learning. The main subject I have worked on is support vector machines (SVM) for data classification/regression. We develop the software LIBSVM, which is one of the most widely used SVM packages.

Recently we are interested in large-scale linear classification (e.g., document classification or natural language processing applications). One of our latest software is LIBLINEAR. It can efficiently train millions of documents.

The major computational task behind machine learning methods is to solve optimization problems. I am interested in large-scale numerical optimization and its applications. A nonlinear optimization tool developed earlier is TRON.