Diffusion-BBO: Diffusion-Based Inverse Modeling for Online Black-Box Optimization

2024-06-30

AI Summary

To address low sample efficiency and frequent generation of invalid designs in online black-box optimization (BBO), this paper proposes the first online inverse modeling paradigm based on conditional diffusion models: it actively samples in the target space to guide design-space generation, thereby avoiding implausible solutions. Methodologically, we introduce an Uncertainty-aware Exploration (UaE) acquisition function that enables uncertainty-driven inverse sampling, accompanied by theoretical guarantees of near-optimality and convergence analysis. Experiments across six scientific discovery tasks demonstrate that our approach significantly outperforms existing online BBO baselines—achieving higher solution quality and superior sample efficiency with fewer function evaluations. This work establishes a novel, interpretable, reliable, and efficient generative optimization paradigm for data-scarce scientific discovery scenarios.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Online black-box optimization (BBO) aims to optimize an objective function by iteratively querying a black-box oracle in a sample-efficient way. While prior studies focus on forward approaches such as Gaussian Processes (GPs) to learn a surrogate model for the unknown objective function, they struggle with steering clear of out-of-distribution and invalid designs in scientific discovery tasks. Recently, inverse modeling approaches that map the objective space to the design space with conditional diffusion models have demonstrated impressive capability in learning the data manifold. However, these approaches proceed in an offline fashion with pre-collected data. How to design inverse approaches for online BBO to actively query new data and improve the sample efficiency remains an open question. In this work, we propose Diffusion-BBO, a sample-efficient online BBO framework leveraging the conditional diffusion model as the inverse surrogate model. Diffusion-BBO employs a novel acquisition function Uncertainty-aware Exploration (UaE) to propose scores in the objective space for conditional sampling. We theoretically prove that Diffusion-BBO with UaE achieves a near-optimal solution for online BBO. We also empirically demonstrate that Diffusion-BBO with UaE outperforms existing online BBO baselines across 6 scientific discovery tasks.

Problem

• Online Black-Box Optimization
• Conditional Diffusion Models
• Efficient Optimization

Innovation

• Diffusion-BBO
• Conditional Diffusion Models
• Uncertainty-aware Exploration