# Bayesian Optimisation Using Product-of-Experts Gaussian Process Models with Uncertainty Calibration

arXiv: https://arxiv.org/abs/2609.15555 · HTML: https://mlsanity.com/paper/2609.15555/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 20 references that are in the map.
- First posted on arXiv 14 Sep 2026.

## Builds on (most cited first, 6 of 6)

- [A Tutorial on Bayesian Optimization](https://mlsanity.com/paper/1807.02811/) (2018)
- [GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration](https://mlsanity.com/paper/1809.11165/) (2018)
- [Scalable Global Optimization via Local Bayesian Optimization](https://mlsanity.com/paper/1910.01739/) (2019)
- [When Gaussian Process Meets Big Data: A Review of Scalable GPs](https://mlsanity.com/paper/1807.01065/) (2018)
- [Transductive Log Opinion Pool of Gaussian Process Experts](https://mlsanity.com/paper/1511.07551/) (2015)
- [Information-Based Calibration of Uncertainty Quantification in Product-of-Experts Gaussian Process Models](https://mlsanity.com/paper/2608.29349/) (2026)

## Abstract (arXiv)

Bayesian optimisation (BO) typically relies on a single global Gaussian process (GP) model as its surrogate model. However, GP regression has cubic computational complexity in the number of training data points, limiting its applicability to large-scale optimisation problems. The product-of-experts Gaussian process model with uncertainty calibration (GP-pro-c) mitigates this limitation by combining multiple local GP experts, enabling improved uncertainty quantification, reduced computational cost, and preservation of global correlations. Despite these desirable properties, the use of GP-pro-c in BO has not been thoroughly studied. This paper introduces BO-pro-c, a Bayesian optimisation algorithm that uses GP-pro-c as its surrogate model, and evaluate its performance across a diverse range of BO settings. Experimental results suggest that BO-pro-c maintains competitive optimisation performance while achieving a 0.9% reduction in simple regret and a 39.4% reduction in computational overhead relative to a BO algorithm based on a single global GP model.