Skip to content

Distributional Nearest Neighbors for distributional table completion.

Notifications You must be signed in to change notification settings

jacobf18/Dist-NN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Dist-NN

Authors: Jacob Feitelberg, Kyuseong Choi, Anish Agarwal, Raaz Dwivedi

Distributional Nearest Neighbors is a framework for solving the distributional matrix completion problem. This is similar to scalar matrix completion, but each entry is a distribution of real numbers instead of a single scalar. We use the Wasserstein metric (in particular, the 2-Wasserstein metric) to determine how close distributions are to each other. We choose to use the Wasserstein metric because it retains geometric information when finding an average in the Wasserstein space. For instance, an average of multiple Gaussian distributions is again Gaussian.

About

Distributional Nearest Neighbors for distributional table completion.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published