Work While They Sleep: Exploiting Evaluation Latency for Fully Bayesian Optimization

2026-10-06

AI Summary

This study addresses the impracticality of fully Bayesian approaches in black-box optimization, where computationally expensive Markov chain Monte Carlo (MCMC) sampling induces prohibitive decision latency. We propose ELF-BO, an algorithm that introduces a novel "sample-while-evaluating" asynchronous mechanism. By exploiting the idle time during objective function evaluations to sample the hyperparameter posterior distribution in parallel, and by incorporating an importance reweighting strategy, ELF-BO entirely conceals the costly MCMC computations within the evaluation latency, thereby eliminating additional decision overhead. Experiments on both synthetic and real-world tasks demonstrate that the proposed method achieves performance comparable to full Bayesian optimization while exhibiting lower decision latency than standard Bayesian optimization. This work renders fully Bayesian optimization practically viable for the first time.

Abstract

Black-box optimization problems are ubiquitous across science and engineering, often dealing with expensive objective functions. This objective latency has two consequences during optimization: (i) the objective evaluation dominates execution time, and (ii) sample-efficient algorithms are crucial to accelerate development and avoid wasting resources. Bayesian optimization (BO) methods are the \textit{de facto} choice of planners for suggesting the next point to try. Standard BO fits the surrogate model's hyperparameters with a point estimate. Alternatively, a fully Bayesian approach uses model averaging to account for uncertainty over the hyperparameters, leading to better uncertainty estimates---useful in the low-data regime that is pervasive in BO. However, it is often prohibitively expensive and thus rarely used. In this work, we propose ELF-BO, an algorithm that uses the objective evaluation latency to headstart the computation of the next suggestion, allowing for fully Bayesian optimization without incurring substantial decision-time costs. This is done by sampling from the hyperparameter posterior \emph{while} the objective is being evaluated, only requiring reweighting of the samples once the objective value is observed. Across synthetic functions and real-world applications, we show that ELF-BO matches the performance of fully Bayesian methods while only incurring decision latency on par with or better than standard BO. Thus, ELF-BO makes fully Bayesian optimization practical in real-world use cases.

Problem

• Black-box optimization
• Bayesian optimization
• Fully Bayesian optimization
• Evaluation latency
• Surrogate model

Innovation

• Fully Bayesian Optimization
• Evaluation Latency
• Black-box Optimization
• Hyperparameter Posterior Sampling
• Sample Reweighting