Accelerating Multi-Objective Bayesian Optimisation via Predictive-Gradient Catalysts

2026-06-05

AI Summary

This work addresses the slow convergence of multi-objective Bayesian optimization (MOBO) under limited evaluation budgets. The authors propose a general acceleration mechanism that, for the first time, incorporates Gaussian process predictive gradients as auxiliary signals into MOBO. By leveraging local smoothness information from the surrogate model without modifying existing Pareto-compliant acquisition functions, the method accelerates convergence toward the global Pareto set. Two catalytic strategies—adaptive and predefined weighting—are introduced to effectively enhance the performance of acquisition functions such as EHVI and AugTch. Empirical results on the DTLZ benchmark suite (2 objectives, 10 dimensions) demonstrate significant improvements over state-of-the-art methods, particularly when the surrogate model is accurate and the underlying problem exhibits sufficient smoothness.

Abstract

This paper presents a general acceleration mechanism for multi-objective Bayesian optimisation (MOBO) that leverages Gaussian process predictive gradients as auxiliary signals. Rather than replacing existing Pareto-compliant acquisition functions, the proposed approach augments them with local stationarity information derived from surrogate-derived gradients, enabling faster convergence toward the global Pareto set under limited evaluation budgets. Two catalyst instantiations are investigated: an adaptive Multiple-Gradient Descent Algorithm-Based Catalyst (MGDA) and a predefined-weight variant that enables focused exploration when budgets are tight. Experiments on the DTLZ benchmark suite (using 2 objectives and 10 decision variables) show that predictive gradient catalysis can deliver significant acceleration compared to other acquisition functions (EHVI, AugTch, tMPoI, SAF) when surrogates are accurate, particularly for stationary problems.

Problem

• Multi-Objective Bayesian Optimisation
• Pareto Set
• Convergence Acceleration
• Limited Evaluation Budgets
• Surrogate Gradients

Innovation

• predictive gradients
• multi-objective Bayesian optimisation
• Pareto set acceleration
• Gaussian process surrogates
• acquisition function augmentation