Robust Bloom Filters for Large MultiLabel Classification Tasks

Moustapha Cisse (1) , Nicolas Usunier (2) , Thierry Artières (1) , Patrick Gallinari (1)

1 MLIA - Machine Learning and Information Access
2 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]

This paper presents an approach to multilabel classification (MLC) with a large number of labels. Our approach is a reduction to binary classification in which label sets are represented by low dimensional binary vectors. This representation follows the principle of Bloom filters, a space-efficient data structure originally designed for approximate membership testing. We show that a naive application of Bloom filters in MLC is not robust to individual binary classifiers' errors. We then present an approach that exploits a specific feature of real-world datasets when the number of labels is large: many labels (almost) never appear together. Our approach is provably robust, has sublinear training and inference complexity with respect to the number of labels, and compares favorably to state-of-the-art algorithms on two large scale multilabel datasets.

Domaines: Apprentissage [cs.LG]

Moustapha Cisse, Nicolas Usunier, Thierry Artières, Patrick Gallinari. Robust Bloom Filters for Large MultiLabel Classification Tasks. Advances in Neural Information Processing Systems 26, Dec 2013, Lake Tahoe, United States. pp.1851-1859. ⟨hal-00942742⟩