Unverified60% confidenceSolutionExact time
Wasserstein GAN replaces the JS divergence with the Wasserstein distance, providing stable gradients even when distributions do not overlap.
2
Sources
60%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
Related Claims
UnverifiedWasserstein GAN (WGAN) was proposed by Martin Arjovsky et al. in 2017, replacing the JS divergence in the original GAN with the Wasserstein distance as the training objective.70% similarUnverifiedUnlike KL divergence, the Wasserstein distance has geometry-aware properties and does not diverge to infinity as KL divergence does when the supports do not completely overlap.68% similarUnverifiedJS divergence is a symmetrized version of KL divergence, with results always bounded in [0, ln2], avoiding KL divergence's asymmetry and the defect of diverging to infinity when support sets do not overlap.66% similarUnverified切片Wasserstein距离的理论基础来自Radon变换，其逆变换的存在性保证从一维投影集合可完整恢复原始高维分布信息58% similarUnverifiedLindgren et al. proved that a Gaussian process with a Matérn covariance function can be represented as the solution of some stochastic partial differential equation (SPDE), and after discretizing it on a triangular mesh with the finite element method, its coefficients form a Gaussian Markov random field (GMRF).56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/563690API
curl https://kongchang.com/api/v1/knowledge/claims/563690MCP
get_claim(id=563690)