Sobolev regularization on the witness function enables global convergence of MMD gradient flows for both sampling and generative modeling without isoperimetric assumptions.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
EDA-specific RHS differences used as sketching matrix for randomized Hessian preconditioner accelerate linear solves across the ensemble in Lorenz-96 experiments.
citing papers explorer
-
Sobolev Regularized MMD Gradient Flow
Sobolev regularization on the witness function enables global convergence of MMD gradient flows for both sampling and generative modeling without isoperimetric assumptions.
-
Accelerating an ensemble of variational data assimilations with randomized preconditioning
EDA-specific RHS differences used as sketching matrix for randomized Hessian preconditioner accelerate linear solves across the ensemble in Lorenz-96 experiments.