The objective in extreme multi-label learning is to learn a classifier that can automatically tag a datapoint with the most relevant subset of labels from an extremely large label set. This competition  provides an extreme classification dataset with 4K labels and 15K training points.