Geometric Losses for Distributional Learning
arXiv:1905.06005
Abstract
Building upon recent advances in entropy-regularized optimal transport, and upon Fenchel duality between measures and continuous functions , we propose a generalization of the logistic loss that incorporates a metric or cost between classes. Unlike previous attempts to use optimal transport distances for learning, our loss results in unconstrained convex objective functions, supports infinite (or very large) class spaces, and naturally defines a geometric generalization of the softmax operator. The geometric properties of this loss make it suitable for predicting sparse and singular distributions, for instance supported on curves or hyper-surfaces. We study the theoretical properties of our loss and show-case its effectiveness on two applications: ordinal regression and drawing generation.
References in corpus (9)
- A Neural Representation of Sketch Drawings
- Fast Computation of Wasserstein Barycenters
- Learning with a Wasserstein Loss
- Composite Binary Losses
- Interpolating between Optimal Transport and MMD using Sinkhorn Divergences
- Differential Properties of Sinkhorn Approximation for Learning with Wasserstein Distance
- Semi-dual Regularized Optimal Transport
- A Consistent Regularization Approach for Structured Prediction
- Learning with Fenchel-Young Losses