---
title: Cost-aware Bayesian optimization via the Pandora's Box Gittins index
url: https://www.ml-quant.com/papers/arxiv/2406.20062/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: arXiv:2406.20062
source_url: https://arxiv.org/abs/2406.20062
featured: 2024-07-03
citations: 22
topic: Econometrics & Forecasting
---


# Cost-aware Bayesian optimization via the Pandora's Box Gittins index

The paper suggests using the Gittins index from the Pandora's Box problem in economics as a function for cost-aware Bayesian optimization.

- Source: https://arxiv.org/abs/2406.20062
- Identifier: arXiv:2406.20062
- Released: 2024-06-28
- First featured: Quant Letter No. 55 (2024-07-03): https://www.ml-quant.com/issues/2024-07-03/
- Citations (Semantic Scholar): 22
- Published in: Neural Information Processing Systems
- Topic: Econometrics & Forecasting

## Related

- [Selecting Experimental Sites for External Validity](https://www.ml-quant.com/papers/arxiv/2405.13241/): A proposed Bayesian decision-theoretic approach optimizes external validity in experiments, revealing efficiency losses when using evidence from randomly-selected sites or those with the largest expected treatment effects.
- [Accelerating Convergence in Bayesian Few-Shot Classification](https://www.ml-quant.com/papers/arxiv/2405.01507/): The research combines mirror descent-based variational inference with Gaussian process for few-shot classification, enhancing accuracy, uncertainty measurement, and faster convergence.
- [Diffusive Gibbs Sampling](https://www.ml-quant.com/papers/arxiv/2402.03008/): The article introduces Diffusive Gibbs Sampling (DiGS), a new method for sampling from multi-modal distributions, which performs better in tasks like Bayesian neural networks and molecular dynamics.
- [Amortized Bayesian Multilevel Models](https://www.ml-quant.com/papers/arxiv/2408.13230/): Research explores the use of neural network architectures in Bayesian Multilevel Models to enable efficient training and inference on unseen data sets, solving computational challenges and providing quick posterior inference.
- [Bayesian Optimization for Non-Convex Two-Stage Stochastic Optimization Problems](https://www.ml-quant.com/papers/arxiv/2408.17387/): The study uses Bayesian optimization to solve complex, two-stage stochastic programs, showing better results than standard methods and providing a more efficient solution to uncertain optimization problems.
- [TS-RSR: A Provably Efficient Approach for Batch Bayesian Optimization](https://www.ml-quant.com/papers/arxiv/2403.04764/): A novel method for batch Bayesian Optimization (BO) is introduced, which reduces redundancy and focuses on points with high predictive means or uncertainty, showing superior performance on nonconvex test functions.