Computationally and Statistically Efficient Inference for Complex Large-scale Data

Abstract

The aim of the highly successful workshop Computationally and statistically efficient inference for large-scale and heterogeneous data was to foster dissemination and collaboration between researchers in the area of highdimensional and large-scale data analysis. The field has grown tremendously over the last decade. Faced with ever larger data sets, many algorithms have emerged in computer science, machine learning and statistics that allow computationally efficient manipulation and model fitting on large datasets. Yet the mathematical and statistical properties of these algorithms are only just beginning to be understood. Advancing the field is important to avoid many misleading scientific discoveries based on pure data manipulation without the accompanying mathematical insights. The talks and discussions at the workshop covered the latest advances from optimization to statistical error control for large-scale data analysis.