Improvements in engineering and data acquisition techniques have rendered large amounts, of potentially high dimensional, data easily available. As a result, statistical analysis of big, high-dimensional data has become frequent in many scientific fields ranging from biology, genomics and health sciences to astronomy, economics and machine learning. Although nonparametric methods are better suitable to model complex systems underlying data generating processes and often achieve state of the art performance on a wide range of tasks, majority of the research in machine learning focuses on linear models. This may come from the common belief that nonparametric methods do not scale to big data problems.
The aim of this workshop is to bring together practitioners, who work on specialized applications, and theoreticians that are interested in providing sound methodology. It is important that we effectively communicate advances arising in different areas of machine learning and drawbacks of existing methods to develop methodology that matters. In particular, we hope to educate theoreticians of needs in real world applications and practitioners of new methodological advances. We will bring together machine learners and statisticians, since both communities work on the topic, but emphasise different aspects of the nonparametric learning and have complementary strengths. Furthermore, we hope to advertise recent successes of nonparametric methods in a number of domains, involving large scale high-dimensional problems, and to dismiss the common belief that nonparametric methods are not suitable for dealing with challenges arising from big data.
We gratefully acknowledge our sponsors.