Support-Proximity Augmented Diffusion Estimation for Offline Black-Box Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Yonghan, Yuan, Ye, Sun, Zipeng, Du, Linfeng, He, Bowei, Wu, Haolun, Chen, Can, Liu, Xue
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913151860080640
author Yang, Yonghan
Yuan, Ye
Sun, Zipeng
Du, Linfeng
He, Bowei
Wu, Haolun
Chen, Can
Liu, Xue
author_facet Yang, Yonghan
Yuan, Ye
Sun, Zipeng
Du, Linfeng
He, Bowei
Wu, Haolun
Chen, Can
Liu, Xue
contents Offline black-box optimization aims to discover novel designs with high property scores using only a static dataset, a task fundamentally challenged by the out-of-distribution (OOD) extrapolation problem. Existing approaches typically bifurcate into inverse methods, which struggle with the ill-posed nature of mapping scores to designs, and forward methods, which often lack the distributional expressivity to quantify uncertainty effectively. In this work, we propose SPADE (Support-Proximity Augmented Diffusion Estimation), a novel framework that reimagines forward surrogate modeling through the lens of conditional generative modeling. SPADE models the forward likelihood p(y|x) using a diffusion model, but with two critical enhancements to tailor it for optimization: (1) a Calibrated Diffusion Estimation module that enforces global consistency in statistical moments and pairwise rankings, and (2) a Support-Proximity Regularization mechanism that implicitly internalizes the data manifold constraint p(x) via kNN-based density estimation. Theoretically, we prove that our regularization is first-order equivalent to maximizing a Bayesian posterior with a valid design prior. Empirically, SPADE achieves state-of-the-art performance across Design-Bench tasks and an LLM data mixture optimization benchmark.
format Preprint
id arxiv_https___arxiv_org_abs_2605_11246
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Support-Proximity Augmented Diffusion Estimation for Offline Black-Box Optimization
Yang, Yonghan
Yuan, Ye
Sun, Zipeng
Du, Linfeng
He, Bowei
Wu, Haolun
Chen, Can
Liu, Xue
Machine Learning
Offline black-box optimization aims to discover novel designs with high property scores using only a static dataset, a task fundamentally challenged by the out-of-distribution (OOD) extrapolation problem. Existing approaches typically bifurcate into inverse methods, which struggle with the ill-posed nature of mapping scores to designs, and forward methods, which often lack the distributional expressivity to quantify uncertainty effectively. In this work, we propose SPADE (Support-Proximity Augmented Diffusion Estimation), a novel framework that reimagines forward surrogate modeling through the lens of conditional generative modeling. SPADE models the forward likelihood p(y|x) using a diffusion model, but with two critical enhancements to tailor it for optimization: (1) a Calibrated Diffusion Estimation module that enforces global consistency in statistical moments and pairwise rankings, and (2) a Support-Proximity Regularization mechanism that implicitly internalizes the data manifold constraint p(x) via kNN-based density estimation. Theoretically, we prove that our regularization is first-order equivalent to maximizing a Bayesian posterior with a valid design prior. Empirically, SPADE achieves state-of-the-art performance across Design-Bench tasks and an LLM data mixture optimization benchmark.
title Support-Proximity Augmented Diffusion Estimation for Offline Black-Box Optimization
topic Machine Learning
url https://arxiv.org/abs/2605.11246