Saved in:
Bibliographic Details
Main Authors: Guerra, Juan D., Garbay, Thomas, Lajoie, Guillaume, Bonizzato, Marco
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.11294
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913997208420352
author Guerra, Juan D.
Garbay, Thomas
Lajoie, Guillaume
Bonizzato, Marco
author_facet Guerra, Juan D.
Garbay, Thomas
Lajoie, Guillaume
Bonizzato, Marco
contents Hierarchical Gaussian Process (H-GP) models divide problems into different subtasks, allowing for different models to address each part, making them well-suited for problems with inherent hierarchical structure. However, typical H-GP models do not fully take advantage of this structure, only sending information up or down the hierarchy. This one-way coupling limits sample efficiency and slows convergence. We propose Bidirectional Information Flow (BIF), an efficient H-GP framework that establishes bidirectional information exchange between parent and child models in H-GPs for online training. BIF retains the modular structure of hierarchical models - the parent combines subtask knowledge from children GPs - while introducing top-down feedback to continually refine children models during online learning. This mutual exchange improves sample efficiency, enables robust training, and allows modular reuse of learned subtask models. BIF outperforms conventional H-GP Bayesian Optimization methods, achieving up to 4x and 3x higher $R^2$ scores for the parent and children respectively, on synthetic and real-world neurostimulation optimization tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2505_11294
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bidirectional Information Flow (BIF) -- A Sample Efficient Hierarchical Gaussian Process for Bayesian Optimization
Guerra, Juan D.
Garbay, Thomas
Lajoie, Guillaume
Bonizzato, Marco
Machine Learning
Hierarchical Gaussian Process (H-GP) models divide problems into different subtasks, allowing for different models to address each part, making them well-suited for problems with inherent hierarchical structure. However, typical H-GP models do not fully take advantage of this structure, only sending information up or down the hierarchy. This one-way coupling limits sample efficiency and slows convergence. We propose Bidirectional Information Flow (BIF), an efficient H-GP framework that establishes bidirectional information exchange between parent and child models in H-GPs for online training. BIF retains the modular structure of hierarchical models - the parent combines subtask knowledge from children GPs - while introducing top-down feedback to continually refine children models during online learning. This mutual exchange improves sample efficiency, enables robust training, and allows modular reuse of learned subtask models. BIF outperforms conventional H-GP Bayesian Optimization methods, achieving up to 4x and 3x higher $R^2$ scores for the parent and children respectively, on synthetic and real-world neurostimulation optimization tasks.
title Bidirectional Information Flow (BIF) -- A Sample Efficient Hierarchical Gaussian Process for Bayesian Optimization
topic Machine Learning
url https://arxiv.org/abs/2505.11294