Emulation of physical processes with Emukit

Andrei Paleyes; Mark Pullin; Maren Mahsereci; Cliff McCollum; Neil D. Lawrence; Javier Gonzalez

edit

Back to publications

Emulation of physical processes with Emukit

Andrei Paleyes, Mark Pullin, Maren Mahsereci, Cliff McCollum, Neil D. Lawrence, Javier Gonzalez

, 2021.

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.

Links

Cite this Paper

BibTeX


@Misc{emulation-of-physical-processes-with-emukit,
  title = 	 {Emulation of physical processes with Emukit},
  author = 	 {Paleyes, Andrei and Pullin, Mark and Mahsereci, Maren and McCollum, Cliff and Lawrence, Neil D. and Gonzalez, Javier},
  year = 	 {2021},
  url = 	 {/publications/emulation-of-physical-processes-with-emukit.html},
  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.
}
}

Endnote

%0 Generic
%T Emulation of physical processes with Emukit
%A Andrei Paleyes
%A Mark Pullin
%A Maren Mahsereci
%A Cliff McCollum
%A Neil D. Lawrence
%A Javier Gonzalez
%D 2021	
%F emulation-of-physical-processes-with-emukit
%U /publications/emulation-of-physical-processes-with-emukit.html
%X 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.

RIS


TY  - GEN
TI  - Emulation of physical processes with Emukit
AU  - Andrei Paleyes
AU  - Mark Pullin
AU  - Maren Mahsereci
AU  - Cliff McCollum
AU  - Neil D. Lawrence
AU  - Javier Gonzalez
DA  - 2021/10/25	
ID  - emulation-of-physical-processes-with-emukit
UR  - /publications/emulation-of-physical-processes-with-emukit.html
AB  - 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.

ER  -

APA


Paleyes, A., Pullin, M., Mahsereci, M., McCollum, C., Lawrence, N.D. & Gonzalez, J.. (2021). Emulation of physical processes with Emukit.  Available from /publications/emulation-of-physical-processes-with-emukit.html.