Summary of A Quadrature Approach For General-purpose Batch Bayesian Optimization Via Probabilistic Lifting, by Masaki Adachi et al.
A Quadrature Approach for General-Purpose Batch Bayesian Optimization via Probabilistic Lifting
by Masaki Adachi, Satoshi Hayakawa, Martin Jørgensen, Saad Hamid, Harald Oberhauser, Michael A. Osborne
First submitted to arxiv on: 18 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Numerical Analysis (math.NA); Machine Learning (stat.ML)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents SOBER, a modular framework for batch Bayesian optimisation via probabilistic lifting with kernel quadrature. This framework addresses challenges in parallelising Bayesian optimisation, including flexibility in acquisition functions and kernel choices, handling discrete and continuous variables simultaneously, model misspecification, and fast massive parallelisation. SOBER offers versatility in downstream tasks under a unified approach, gradient-free sampling for domain-agnostic sampling, flexibility in domain prior distribution, adaptive batch size determination, robustness against misspecified reproducing kernel Hilbert space, and natural stopping criterion. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper makes it easier to use Bayesian optimisation, which is a way to find the best combination of variables. It solves some big problems with this method, like being able to handle different types of data and making sure the results are good even if the assumptions aren’t correct. The new approach is called SOBER, and it lets you do many things at once, like sampling without knowing the gradient of a function. This makes it useful for lots of different problems and helps ensure that the results are reliable. |
