SPADE is a conditional diffusion model for offline black-box optimization that adds calibrated moment/ranking consistency and kNN support-proximity regularization, with a claimed first-order Bayesian equivalence.
Proceedings of The 33rd International Conference on Machine Learning , publisher =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Support-Proximity Augmented Diffusion Estimation for Offline Black-Box Optimization
SPADE is a conditional diffusion model for offline black-box optimization that adds calibrated moment/ranking consistency and kNN support-proximity regularization, with a claimed first-order Bayesian equivalence.