PI is back! Switching Acquisition Functions in Bayesian Optimization

by   Carolin Benjamins, et al.

Bayesian Optimization (BO) is a powerful, sample-efficient technique to optimize expensive-to-evaluate functions. Each of the BO components, such as the surrogate model, the acquisition function (AF), or the initial design, is subject to a wide range of design choices. Selecting the right components for a given optimization task is a challenging task, which can have significant impact on the quality of the obtained results. In this work, we initiate the analysis of which AF to favor for which optimization scenarios. To this end, we benchmark SMAC3 using Expected Improvement (EI) and Probability of Improvement (PI) as acquisition functions on the 24 BBOB functions of the COCO environment. We compare their results with those of schedules switching between AFs. One schedule aims to use EI's explorative behavior in the early optimization steps, and then switches to PI for a better exploitation in the final steps. We also compare this to a random schedule and round-robin selection of EI and PI. We observe that dynamic schedules oftentimes outperform any single static one. Our results suggest that a schedule that allocates the first 25 optimization budget to EI and the last 75 However, we also observe considerable performance differences for the 24 functions, suggesting that a per-instance allocation, possibly learned on the fly, could offer significant improvement over the state-of-the-art BO designs.


page 4

page 7

page 16

page 19


Towards Automated Design of Bayesian Optimization via Exploratory Landscape Analysis

Bayesian optimization (BO) algorithms form a class of surrogate-based he...

Efficient Rollout Strategies for Bayesian Optimization

Bayesian optimization (BO) is a class of sample-efficient global optimiz...

Revisiting Bayesian Optimization in the light of the COCO benchmark

It is commonly believed that Bayesian optimization (BO) algorithms are h...

Comparison of High-Dimensional Bayesian Optimization Algorithms on BBOB

Bayesian Optimization (BO) is a class of black-box, surrogate-based heur...

AutoLRS: Automatic Learning-Rate Schedule by Bayesian Optimization on the Fly

The learning rate (LR) schedule is one of the most important hyper-param...

Self-Adjusting Weighted Expected Improvement for Bayesian Optimization

Bayesian Optimization (BO) is a class of surrogate-based, sample-efficie...

Expected Improvement versus Predicted Value in Surrogate-Based Optimization

Surrogate-based optimization relies on so-called infill criteria (acquis...

Please sign up or login with your details

Forgot password? Click here to reset