Events Detail

Speaker:
Prof. Dr. Johannes Lederer, Universität Bochum

Abstract:

Sparsity is popular in statistics and machine learning, because it can avoid overfitting, speed up computations, and facilitate interpretations in deep learning, however, the full potential of sparsity still needs to be explored In this presentation, we first discuss sparsity in the framework of high dimensional statistics and corresponding statistical theories. We then use these insights to further our understanding of sparsity in deep learning.