Emulation of physical processes with Emukit

TL;DR

Emukit is a versatile Python toolkit that integrates advanced decision-making methods like Bayesian optimization and emulation, enabling flexible and customizable uncertainty quantification for complex systems.

Contribution

It introduces Emukit, a unified, adaptable toolkit that consolidates various decision-making and emulation techniques, supporting custom models and easy prototyping.

Findings

Demonstrated Emukit's application on three case studies.

Showcased integration of multiple decision-making methods.

Validated flexibility and extensibility of the toolkit.

Abstract

Decision making in uncertain scenarios is an ubiquitous challenge in real world systems. Tools to deal with this challenge include simulations to gather information and statistical emulation to quantify uncertainty. The machine learning community has developed a number of methods to facilitate decision making, but so far they are scattered in multiple different toolkits, and generally rely on a fixed backend. In this paper, we present Emukit, a highly adaptable Python toolkit for enriching decision making under uncertainty. Emukit allows users to: (i) use state of the art methods including Bayesian optimization, multi-fidelity emulation, experimental design, Bayesian quadrature and sensitivity analysis; (ii) easily prototype new decision making methods for new problems. Emukit is agnostic to the underlying modeling framework and enables users to use their own custom models. We show how Emukit can be used on three exemplary case studies.