Learning Surrogates for Offline Black-Box Optimization via Gradient Matching

2025-02-26

AI Summary

In offline black-box optimization, surrogate models exhibit poor predictive accuracy in out-of-distribution regions, leading to a substantial performance gap between the surrogate-optimal and true-optimal solutions. To address this, we establish—for the first time—a theoretical bound linking the fidelity of surrogate gradient field alignment with the optimization performance gap. Building on this insight, we propose Gradient Matching Learning (GML), a novel paradigm that jointly performs implicit gradient field modeling and gradient-matching regularization, yielding provably improved optimization quality. Evaluated on multiple real-world design benchmarks from materials science and chemistry, GML consistently outperforms state-of-the-art methods, significantly narrowing the performance gap between surrogate-optimal and true-optimal solutions. Our framework thus provides a theoretically grounded and empirically effective surrogate optimization paradigm for high-cost experimental settings.

Technology Category

• Machine Learning: Optimization
• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations

Abstract

Offline design optimization problem arises in numerous science and engineering applications including material and chemical design, where expensive online experimentation necessitates the use of in silico surrogate functions to predict and maximize the target objective over candidate designs. Although these surrogates can be learned from offline data, their predictions are often inaccurate outside the offline data regime. This challenge raises a fundamental question about the impact of imperfect surrogate model on the performance gap between its optima and the true optima, and to what extent the performance loss can be mitigated. Although prior work developed methods to improve the robustness of surrogate models and their associated optimization processes, a provably quantifiable relationship between an imperfect surrogate and the corresponding performance gap, as well as whether prior methods directly address it, remain elusive. To shed light on this important question, we present a theoretical framework to understand offline black-box optimization, by explicitly bounding the optimization quality based on how well the surrogate matches the latent gradient field that underlines the offline data. Inspired by our theoretical analysis, we propose a principled black-box gradient matching algorithm to create effective surrogate models for offline optimization, improving over prior approaches on various real-world benchmarks.

Problem

• Impact of imperfect surrogate models on optimization performance gap.
• Quantifiable relationship between surrogate inaccuracy and performance loss.
• Development of gradient matching algorithm for effective surrogate models.

Innovation

• Gradient matching for surrogate models
• Theoretical framework for optimization quality
• Black-box gradient matching algorithm