Salvato in:
Dettagli Bibliografici
Autori principali: Hong, Shu, Mei, Yongsheng, Imani, Mahdi, Lan, Tian
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2511.07734
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914149997477888
author Hong, Shu
Mei, Yongsheng
Imani, Mahdi
Lan, Tian
author_facet Hong, Shu
Mei, Yongsheng
Imani, Mahdi
Lan, Tian
contents Bayesian optimization (BO) is a powerful framework for optimizing expensive black-box objectives, yet extending it to graph-structured domains remains challenging due to the discrete and combinatorial nature of graphs. Existing approaches often rely on either full graph topology-impractical for large or partially observed graphs-or incremental exploration, which can lead to slow convergence. We introduce a scalable framework for global optimization over graphs that employs low-rank spectral representations to build Gaussian process (GP) surrogates from sparse structural observations. The method jointly infers graph structure and node representations through learnable embeddings, enabling efficient global search and principled uncertainty estimation even with limited data. We also provide theoretical analysis establishing conditions for accurate recovery of underlying graph structure under different sampling regimes. Experiments on synthetic and real-world datasets demonstrate that our approach achieves faster convergence and improved optimization performance compared to prior methods.
format Preprint
id arxiv_https___arxiv_org_abs_2511_07734
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Global Optimization on Graph-Structured Data via Gaussian Processes with Spectral Representations
Hong, Shu
Mei, Yongsheng
Imani, Mahdi
Lan, Tian
Machine Learning
Artificial Intelligence
Social and Information Networks
Signal Processing
Bayesian optimization (BO) is a powerful framework for optimizing expensive black-box objectives, yet extending it to graph-structured domains remains challenging due to the discrete and combinatorial nature of graphs. Existing approaches often rely on either full graph topology-impractical for large or partially observed graphs-or incremental exploration, which can lead to slow convergence. We introduce a scalable framework for global optimization over graphs that employs low-rank spectral representations to build Gaussian process (GP) surrogates from sparse structural observations. The method jointly infers graph structure and node representations through learnable embeddings, enabling efficient global search and principled uncertainty estimation even with limited data. We also provide theoretical analysis establishing conditions for accurate recovery of underlying graph structure under different sampling regimes. Experiments on synthetic and real-world datasets demonstrate that our approach achieves faster convergence and improved optimization performance compared to prior methods.
title Global Optimization on Graph-Structured Data via Gaussian Processes with Spectral Representations
topic Machine Learning
Artificial Intelligence
Social and Information Networks
Signal Processing
url https://arxiv.org/abs/2511.07734