DeepAI AI Chat

Wasserstein Neural Processes

10/01/2019

∙

by Andrew Carr, et al.

∙

∙

Neural Processes (NPs) are a class of models that learn a mapping from a context set of input-output pairs to a distribution over functions. They are traditionally trained using maximum likelihood with a KL divergence regularization term. We show that there are desirable classes of problems where NPs, with this loss, fail to learn any reasonable distribution. We also show that this drawback is solved by using approximations of Wasserstein distance which calculates optimal transport distances even for distributions of disjoint support. We give experimental justification for our method and demonstrate performance. These Wasserstein Neural Processes (WNPs) maintain all of the benefits of traditional NPs while being able to approximate a new class of function mappings.

Andrew Carr
2 publications
Jared Nielson
1 publication
David Wingate
19 publications

∙ 10/04/2019

Fused Gromov-Wasserstein Alignment for Hawkes Processes

We propose a novel fused Gromov-Wasserstein alignment method to jointly ...

0 Dixin Luo, et al. ∙

∙ 02/12/2018

A note on reinforcement learning with Wasserstein distance regularisation, with applications to multipolicy learning

In this note we describe an application of Wasserstein distance to Reinf...

0 Mohammed Amin Abdullah, et al. ∙

∙ 07/26/2022

Sliced Wasserstein Variational Inference

Variational Inference approximates an unnormalized distribution via the ...

0 Mingxuan Yi, et al. ∙

∙ 03/17/2022

Low-rank Wasserstein polynomial chaos expansions in the framework of optimal transport

A unsupervised learning approach for the computation of an explicit func...

0 Robert Gruhlke, et al. ∙

∙ 11/19/2020

Wasserstein Learning of Determinantal Point Processes

Determinantal point processes (DPPs) have received significant attention...

0 Lucas Anquetil, et al. ∙

∙ 02/15/2018

Stochastic Wasserstein Barycenters

We present a stochastic algorithm to compute the barycenter of a set of ...

0 Sebastian Claici, et al. ∙

∙ 03/08/2018

Distributed Computation of Wasserstein Barycenters over Networks

We propose a new class-optimal algorithm for the distributed computation...

0 César A. Uribe, et al. ∙