Luise, G; Rudi, A; Pontil, M; Ciliberto, C; (2018) Differential Properties of Sinkhorn Approximation for Learning with Wasserstein Distance. In: Bengio, S and Wallach, H and Larochelle, H and Grauman, K and CesaBianchi, N and Garnett, R, (eds.) Advances in Neural Information Processing Systems 31 (NIPS 2018). Neural Information Processing Systems Foundation, Inc.: Montréal, Canada.

Preview

Text Pontil_Differential Properties of Sinkhorn Approximation for Learning with Wasserstein Distance_VoR.pdf - Published Version Download (676kB) | Preview

Abstract

Applications of optimal transport have recently gained remarkable attention thanks to the computational advantages of entropic regularization. However, in most situations the Sinkhorn approximation of the Wasserstein distance is replaced by a regularized version that is less accurate but easy to differentiate. In this work we characterize the differential properties of the original Sinkhorn distance, proving that it enjoys the same smoothness as its regularized version and we explicitly provide an efficient algorithm to compute its gradient. We show that this result benefits both theory and applications: on one hand, high order smoothness confers statistical guarantees to learning with Wasserstein approximations. On the other hand, the gradient formula allows us to efficiently solve learning and optimization problems in practice. Promising preliminary experiments complement our analysis.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item