Asymptotic Guarantees for Learning Generative Models with the Sliced-Wasserstein Distance

Kimia Nadjahi, Alain Durmus, Umut Simsekli, Roland Badeau

Feb-14-2026, 02:45:59 GMT–Neural Information Processing Systems 

In practice,θ is typically chosen as a neural network, for which it is often impossible to evaluate the induced densityµθ.