Generalized Sliced Wasserstein Distances
read the original abstract
The Wasserstein distance and its variations, e.g., the sliced-Wasserstein (SW) distance, have recently drawn attention from the machine learning community. The SW distance, specifically, was shown to have similar properties to the Wasserstein distance, while being much simpler to compute, and is therefore used in various applications including generative modeling and general supervised/unsupervised learning. In this paper, we first clarify the mathematical connection between the SW distance and the Radon transform. We then utilize the generalized Radon transform to define a new family of distances for probability measures, which we call generalized sliced-Wasserstein (GSW) distances. We also show that, similar to the SW distance, the GSW distance can be extended to a maximum GSW (max-GSW) distance. We then provide the conditions under which GSW and max-GSW distances are indeed distances. Finally, we compare the numerical performance of the proposed distances on several generative modeling tasks, including SW flows and SW auto-encoders.
This paper has not been read by Pith yet.
Forward citations
Cited by 4 Pith papers
-
Neural Scaling Laws for Jet Generation
Scaling laws hold logarithmically for model size in autoregressive jet generation, with next-token loss correlating to physical metrics via sliced Wasserstein distance, but show weaker scaling for dataset size and com...
-
Population-Aware Physics-Informed Neural Particle Flow for Bayesian Update
PA-PINPF adds Deep Sets population encoders (state or feature) to PINPF for better Bayesian posterior particle transport on range-measurement and TDOA tasks.
-
Posterior sampling in the Age of Emulators
Compares MH, MALA, HMC, NUTS, and AIES on differentiable likelihood emulators for ΛCDM and sterile-neutrino models, finding MALA and MH competitive in wall time despite NUTS needing fewer samples.
-
Representation Without Reward: A JEPA Audit for LLM Fine-Tuning
An empirical audit of 22 JEPA-style training auxiliaries on Llama-3.2-1B fine-tuning for regex generation finds no statistically significant task improvement after multiple-testing correction, even when auxiliaries vi...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.