REVIEW 2 cited by
An Entropy Search Portfolio for Bayesian Optimization
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
An Entropy Search Portfolio for Bayesian Optimization
read the original abstract
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.
Forward citations
Cited by 2 Pith papers
-
CARE: Controlling LLM-Generated Policies through Auditable Review of Evidence in Scientific Experimentation
CARE improves final-best performance on two HTE benchmarks by using an evidence gate to control LLM policy revisions while defaulting to a standard optimizer.
-
CARE: Controlling LLM-Generated Policies through Auditable Review of Evidence in Scientific Experimentation
CARE introduces an evidence-gated auditable controller for LLM-generated ranking policies in high-throughput experimentation, reporting performance gains on Minerva/Olympus (80.0 to 88.5) and ChemLex (83.9 to 92.1) be...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.