Unbalanced Sinkhorn divergences

[zsteve]

Recently came across a use case for this. Might be something worth implementing in the future.
https://arxiv.org/pdf/1910.12958.pdf

[davibarreira]

Very nice paper! Added on to my "to-read" pile. I still have to go through the Sinkhorn code here to fully understand it. I've only done naive implementations, so I have to better understand all the optimizations you guys have done.