Jul 20, 2020 · In this paper, we consider the problem of computing the barycenter of a set of probability distributions under the Sinkhorn divergence.
In this paper, we consider the problem of computing the barycenter of a set of probability distributions under the Sinkhorn divergence.
The main contributions are the following: 1) a new method called Sinkhorn Descent (SD) for solving Sinkhorn barycenter problem based on finding a pushforward ...
It is proved that SD converges to a stationary point at a sublinear rate, and under reasonable assumptions, it is shown that it asymptotically finds a ...
Jul 20, 2020 · In this paper, we consider the problem of computing the barycenter of a set of probability distributions under the Sinkhorn divergence.
Oct 27, 2020 · Sinkhorn Barycenter via Functional Gradient Descent. Accepted Papers by Program Chairs • Sinkhorn Barycenter via Functional Gradient Descent.
This repository contains two projects: Sinkhorn Barycenter via Functional Gradient Descent. To run SD on MNIST, please first run MNIST_to_measure.py for data ...
My current research focuses on developing neural network-based methods for solving partial differential equations using entropy dissipation principles.
People also ask
What is functional gradient descent?
Does gradient descent guarantee you will find the global minima?
Sinkhorn Barycenter via Functional Gradient Descent. To run SD on MNIST, please first run MNIST_to_measure.py for data preprocessing. Dependencies.
Dec 6, 2020 · Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes ...