English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Variational Autoencoder based Bayesian Optimization (VAE-BO) has demonstratedits excellent performance in addressing high-dimensional structuredoptimization problems. However, current mainstream methods overlook thepotential of utilizing a pool of unlabeled data to construct the latent space,while only concentrating on designing sophisticated models to leverage thelabeled data. Despite their effective usage of labeled data, these methodsoften require extra network structures, additional procedure, resulting incomputational inefficiency. To address this issue, we propose a novel method toeffectively utilize unlabeled data with the guidance of labeled data.Specifically, we tailor the pseudo-labeling technique from semi-supervisedlearning to explicitly reveal the relative magnitudes of optimization objectivevalues hidden within the unlabeled data. Based on this technique, we assignappropriate training weights to unlabeled data to enhance the construction of adiscriminative latent space. Furthermore, we treat the VAE encoder and theGaussian Process (GP) in Bayesian optimization as a unified deep kernellearning process, allowing the direct utilization of labeled data, which weterm as Gaussian Process guidance. This directly and effectively integrates thegoal of improving GP accuracy into the VAE training, thereby guiding theconstruction of the latent space. The extensive experiments demonstrate thatour proposed method outperforms existing VAE-BO algorithms in variousoptimization scenarios. Our code will be published athttps://github.com/TaicaiChen/PG-LBO.

PG-LBO: Enhancing High-Dimensional Bayesian Optimization with Pseudo-Label and Gaussian Process Guidance