No Cover Image

Conference Paper/Proceeding/Abstract 724 views 208 downloads

On Bayesian Search for the Feasible Space Under Computationally Expensive Constraints

Alma Rahat Orcid Logo, Michael Wood

Machine Learning, Optimization, and Data Science, Volume: 12566, Pages: 529 - 540

Swansea University Author: Alma Rahat Orcid Logo

Check full text

DOI (Published version): 10.1007/978-3-030-64580-9_44

Abstract

We are often interested in identifying the feasible subset of a decision space under multiple constraints to permit effective design exploration. If determining feasibility required computationally expensive simulations, the cost of exploration would be prohibitive. Bayesian search is data-efficient...

Full description

Published in: Machine Learning, Optimization, and Data Science
ISBN: 9783030645793 9783030645809
ISSN: 0302-9743 1611-3349
Published: Cham Springer International Publishing 2021
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa55060
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract: We are often interested in identifying the feasible subset of a decision space under multiple constraints to permit effective design exploration. If determining feasibility required computationally expensive simulations, the cost of exploration would be prohibitive. Bayesian search is data-efficient for such problems: starting from a small dataset, the central concept is to use Bayesian models of constraints with an acquisition function to locate promising solutions that may improve predictions of feasibility when the dataset is augmented. At the end of this sequential active learning approach with a limited number of expensive evaluations, the models can accurately predict the feasibility of any solution obviating the need for full simulations. In this paper, we propose a novel acquisition function that combines the probability that a solution lies at the boundary between feasible and infeasible spaces (representing exploitation) and the entropy in predictions (representing exploration). Experiments confirmed the efficacy of the proposed function.
Keywords: Active learning; Feasible region; Feasible design exploration; Gaussian processes; Constrained problems
College: Faculty of Science and Engineering
Start Page: 529
End Page: 540

Similar Items