Improving sample efficiency of high dimensional Bayesian optimization with MCMC
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913186671755264 |
|---|---|
| author | Yi, Zeji Wei, Yunyue Cheng, Chu Xin He, Kaibo Sui, Yanan |
| author_facet | Yi, Zeji Wei, Yunyue Cheng, Chu Xin He, Kaibo Sui, Yanan |
| contents | Sequential optimization methods are often confronted with the curse of dimensionality in high-dimensional spaces. Current approaches under the Gaussian process framework are still burdened by the computational complexity of tracking Gaussian process posteriors and need to partition the optimization problem into small regions to ensure exploration or assume an underlying low-dimensional structure. With the idea of transiting the candidate points towards more promising positions, we propose a new method based on Markov Chain Monte Carlo to efficiently sample from an approximated posterior. We provide theoretical guarantees of its convergence in the Gaussian process Thompson sampling setting. We also show experimentally that both the Metropolis-Hastings and the Langevin Dynamics version of our algorithm outperform state-of-the-art methods in high-dimensional sequential optimization and reinforcement learning benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_02650 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Improving sample efficiency of high dimensional Bayesian optimization with MCMC Yi, Zeji Wei, Yunyue Cheng, Chu Xin He, Kaibo Sui, Yanan Machine Learning Sequential optimization methods are often confronted with the curse of dimensionality in high-dimensional spaces. Current approaches under the Gaussian process framework are still burdened by the computational complexity of tracking Gaussian process posteriors and need to partition the optimization problem into small regions to ensure exploration or assume an underlying low-dimensional structure. With the idea of transiting the candidate points towards more promising positions, we propose a new method based on Markov Chain Monte Carlo to efficiently sample from an approximated posterior. We provide theoretical guarantees of its convergence in the Gaussian process Thompson sampling setting. We also show experimentally that both the Metropolis-Hastings and the Langevin Dynamics version of our algorithm outperform state-of-the-art methods in high-dimensional sequential optimization and reinforcement learning benchmarks. |
| title | Improving sample efficiency of high dimensional Bayesian optimization with MCMC |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2401.02650 |