[Forum SIS] StEering Webinar by MHY TAN - City Univ. Hong Kong

[Rossella Berni]

Dear All,
we are glad to announce the first 2021 Webinar under the activities of the
StEering Inter-University Research Center (https://www.steering.academy/).
The webinar will be delivered by Professor Matthias H.Y. Tan, School of
Data Science, City University of Hong Kong. It will take place on Monday,
May 10th, 2021, timetable 10.30am - 12.30am CET (start time 10.30am).
Title, abstract and short biography follow.

The link to join the webinar will be made available by email on May 7th,
2021. Please feel free to distribute this announcement among your
professional community.

IMPORTANT: The webinar is free of charge. We kindly require you to register
by sending an email to centro.steering a disia.unifi.it with
"Registration-Webinar-Prof. Tan" in the subject line. Please also note that
the number of participants is limited to a maximum of 100 participants.

                  Thank you for your kind attention to the above,
                  Best Regards,
                  Rossella Berni
                  StEering Inter-University Research Center

               §§§§§§

Title: Bayesian Optimization of Expected Quadratic Loss for Multiresponse
Computer Experiments with Internal Noise

Abstract: Design of systems based on computer simulations is prevalent. An
important idea to improve design quality, called robust parameter design
(RPD), is to optimize control factors based on the expectation of a loss
function so that the design is robust to noise factor variations. When
computer simulations are time consuming, optimizing the simulator based on
a Gaussian process (GP) emulator for the response is a computationally
efficient approach. For this purpose, acquisition functions (AFs) are used
to sequentially determine the next design point so that the GP emulator can
more accurately locate the optimal setting.
Despite this, few articles consider AFs for positive definite quadratic
forms such as the expected quadratic loss (EQL) function, which is the
standard expected loss function for RPD with nominally-the-best responses.
This paper proposes new AFs for optimizing the EQL, analyzes their
convergence, and develops quick and accurate methods based on the
characteristic function of the EQL to compute them. We apply the AFs to RPD
problems with internal noise factors based on a GP model and an initial
design tailored for such problems. Numerical results indicate that all four
AFs considered have similar performance, and they outperform an
optimization approach based on modeling the quadratic loss as a GP and
maximin Latin hypercube designs.

Short Biography: Matthias Hwai Yong Tan is an associate professor in the
School of Data Science at City University of Hong Kong. He received his
B.Eng. degree in Mechanical-Industrial Engineering from the Universiti
Teknologi Malaysia, an M.Eng. degree in Industrial and Systems Engineering
from the National University of Singapore and a Ph.D. degree in Industrial
and Systems Engineering from Georgia Institute of Technology. His research
interests include uncertainty quantification, design and analysis of
computer experiments, and applied statistics.






-- 


*____________________________________________________________*

* Rossella Berni   PhD*


*Professor of Economic Statistics*

https://www.disia.unifi.it/p-doc2-2017-200*052-B-3f2a3d2f33292d-0.html*
<https://www.disia.unifi.it/p-doc2-2017-200052-B-3f2a3d2f33292d-0.html>
<http://goog_547437445>

http://local.disia.unifi.it/berni


*Centro di Ricerca Interuniversitario - StEering: design, quality and
reliability*

https://www.disia.unifi.it/p186.html
-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20210331/536d6619/attachment.html>