It's concerned with the methodologies of Wasserstein distance calculation.

There are some ways to calculate Wasserstein distance, such as

  • Sinkhorn iteration aided method [1],
  • Neural networks applied method (proposed in WGANs[2]), etc.

As the original Optimal Transport is too hard to solve, those methodologies get feasible thanks for approximations. In this way, the efficiency, the accuracy and the trade-off between them do matter.

  • Which one is the most reasonable way?
  • Is there any paper which has already investigated such an issue?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.