Screening Sinkhorn Algorithm for Regularized Optimal Transport

Mokhtar Z. Alaya (1) , Maxime Berar (2) , Gilles Gasso (1) , Alain Rakotomamonjy (2)

1 LITIS - Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes
2 DocApp - LITIS - Equipe Apprentissage

We introduce in this paper a novel strategy for efficiently approximating the Sinkhorn distance between two discrete measures. After identifying neglectable components of the dual solution of the regularized Sinkhorn problem, we propose to screen those components by directly setting them at that value before entering the Sinkhorn problem. This allows us to solve a smaller Sinkhorn problem while ensuring approximation with provable guarantees. More formally, the approach is based on a new formulation of dual of Sinkhorn divergence problem and on the KKT optimality conditions of this problem, which enable identification of dual components to be screened. This new analysis leads to the Screenkhorn algorithm. We illustrate the efficiency of Screenkhorn on complex tasks such as dimensionality reduction and domain adaptation involving regularized optimal transport.

Mots clés: Domain adaptation; Sinkhorn divergence; Optimal transport; Wasserstein discriminat analysis; Screening

Domaines: Machine Learning [stat.ML]

Mokhtar Z. Alaya, Maxime Berar, Gilles Gasso, Alain Rakotomamonjy. Screening Sinkhorn Algorithm for Regularized Optimal Transport. Advances in Neural Information Processing Systems, 2019, Vancouver, Canada. ⟨hal-02142661v2⟩