# An Entropy Search Portfolio for Bayesian Optimization

arXiv: https://arxiv.org/abs/1406.4625 · HTML: https://mlsanity.com/paper/1406.4625/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Jun 2014.

## Built on by (most cited first, 11 of 11)

- [Transfer Learning for Bayesian Optimization: A Survey](https://mlsanity.com/paper/2302.05927/) (2023)
- [Surrogate modeling for Bayesian optimization beyond a single Gaussian process](https://mlsanity.com/paper/2205.14090/) (2022)
- [An Efficient Batch Constrained Bayesian Optimization Approach for Analog Circuit Synthesis via Multi-objective Acquisition Ensemble](https://mlsanity.com/paper/2106.15412/) (2021)
- [Adaptive Acquisition Selection for Bayesian Optimization with Large Language Models](https://mlsanity.com/paper/2602.07904/) (2026)
- [Enhanced Bayesian Optimization via Preferential Modeling of Abstract Properties](https://mlsanity.com/paper/2402.17343/) (2024)
- [Weighted Ensembles for Active Learning with Adaptivity](https://mlsanity.com/paper/2206.05009/) (2022)
- [Robust Bayesian optimization with reinforcement learned acquisition functions](https://mlsanity.com/paper/2210.00476/) (2022)
- [Information-theoretic Bayesian Optimization: Survey and Tutorial](https://mlsanity.com/paper/2502.06789/) (2025)
- [Combining Bayesian Inference and Reinforcement Learning for Agent Decision Making: A Review](https://mlsanity.com/paper/2505.07911/) (2025)
- [Applications of Gaussian Processes at Extreme Lengthscales: From Molecules to Black Holes](https://mlsanity.com/paper/2303.14291/) (2023)
- [Portfolio-Based Constrained Multi-Objective Bayesian Optimization for Materials Design](https://mlsanity.com/paper/2609.19550/) (2026)

## Abstract (arXiv)

Bayesian optimization is a sample-efficient method for black-box global optimization. How- ever, the performance of a Bayesian optimization method very much depends on its exploration strategy, i.e. the choice of acquisition function, and it is not clear a priori which choice will result in superior performance. While portfolio methods provide an effective, principled way of combining a collection of acquisition functions, they are often based on measures of past performance which can be misleading. To address this issue, we introduce the Entropy Search Portfolio (ESP): a novel approach to portfolio construction which is motivated by information theoretic considerations. We show that ESP outperforms existing portfolio methods on several real and synthetic problems, including geostatistical datasets and simulated control tasks. We not only show that ESP is able to offer performance as good as the best, but unknown, acquisition function, but surprisingly it often gives better performance. Finally, over a wide range of conditions we find that ESP is robust to the inclusion of poor acquisition functions.