13 datasets found

SoLar: a novel Optimal Transport-based framework for partial-label learning, allowing to refine the disambiguated labels towards matching the marginal class prior distribution.

Zero-shot models suffer due to artifacts inherited from pretraining. A particularly detrimental artifact, caused by unbalanced web-scale pretraining data, is mismatched label...

Optimal transport tools (OTT) is a python toolbox that can solve optimal transport problems between point clouds and histograms.

The dataset used in the paper is not explicitly described, but it is mentioned that the authors used a RealNVP architecture to learn an approximate optimal transport map between...

The dataset used in the paper is a generative model, specifically a Wasserstein GAN and a Wasserstein VAE.

The dataset used in the paper is a noise distribution and a high-dimensional data distribution.

The dataset used in the paper is not explicitly described, but it is mentioned that the authors used a high-dimensional continuous distribution p0, p1 for which the ground truth...

The dataset used in the paper is not explicitly described, but it is mentioned that the authors used a 2D setup to illustrate the proof-of-concept of their Optimal Flow Matching...

Conditional score-based diffusion model for unpaired or partially paired training dataset

The dataset is used to test existing neural (continuous) solvers for the Entropic Optimal Transport and Schrödinger Bridge problems.

Curriculum Reinforcement Learning using Optimal Transport via Gradual Domain Adaptation

The dataset is not explicitly described in the paper, but it is mentioned that the authors used a synthetic example to demonstrate the issue with the envelop theorem.

The dataset is used to test the performance of WGAN dual OT solvers.