Batch Bayesian Optimisation via Density-ratio Estimation with Guarantees
Abstract
We extend BORE — a Bayesian optimization framework that recasts the acquisition function as a probabilistic classification problem via density-ratio estimation — to the batch setting, where multiple candidates are evaluated in parallel. We characterize the conditions under which the resulting algorithm enjoys theoretical convergence guarantees and demonstrate its practical effectiveness on a range of black-box optimization benchmarks.
Type
Publication
Advances in Neural Information Processing Systems 35 (NeurIPS2022)
Bayesian Optimization
Density Ratio Estimation
Probabilistic Models
AutoML
Hyperparameter Optimization
Machine Learning
Authors
Research Scientist
My name is Louis Tiao, and I graduated from one of Australia’s
top engineering schools with really good grades.
Now, I’m using my knowledge to help up-and-coming tech companies
make it in this competitive world.