PDNS decomposes diffusion neural sampler training into proximal subproblems on path measures to gradually approach multimodal targets and promote mode exploration.
Z T 0 1 2 ∥ut(Xt)−σ t∇logP ∗ T|t (XT |Xt)∥2dt # .(36) Finally, we obtain theweighted denoising cross entropyoptimization problem: u∗ = argmin u EX∼P ∗
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Proximal Diffusion Neural Sampler
PDNS decomposes diffusion neural sampler training into proximal subproblems on path measures to gradually approach multimodal targets and promote mode exploration.