Physics-integrated segmented Gaussian process (SegGP) learning for cost-efficient training of diesel engine control system with low cetane numbers

SR Narayanan, Y Ji, HD Sapra, S Yang… - AIAA SCITECH 2023 …, 2023 - arc.aiaa.org
View Video Presentation: https://doi. org/10.2514/6.2023-1283. vid Control model training is
an essential step towards the development of an engine controls system. A robust controls …

A hierarchical expected improvement method for bayesian optimization

Z Chen, S Mak, CFJ Wu - Journal of the American Statistical …, 2024 - Taylor & Francis
Abstract The Expected Improvement (EI) method, proposed by Jones, Schonlau, andWelch,
is a widely used Bayesian optimization method, which makes use of a fitted Gaussian …

Efficient emulation of relativistic heavy ion collisions with transfer learning

D Liyanage, Y Ji, D Everett, M Heffernan, U Heinz… - Physical Review C, 2022 - APS
Measurements from the Large Hadron Collider (LHC) and the Relativistic Heavy Ion Collider
(RHIC) can be used to study the properties of quark-gluon plasma. Systematic constraints on …

APIK: Active physics-informed kriging model with partial differential equations

J Chen, Z Chen, C Zhang, CF Jeff Wu - SIAM/ASA Journal on Uncertainty …, 2022 - SIAM
Kriging (or Gaussian process regression) becomes a popular machine learning method for
its flexibility and closed-form prediction expressions. However, one of the key challenges in …

Federated multi-output Gaussian processes

S Chung, R Al Kontar - Technometrics, 2024 - Taylor & Francis
Multi-output Gaussian process (MGP) regression plays an important role in the integrative
analysis of different but interrelated systems/units. Existing MGP approaches assume that …

Additive Multi-Index Gaussian process modeling, with application to multi-physics surrogate modeling of the quark-gluon plasma

K Li, S Mak, JF Paquet, SA Bass - arXiv preprint arXiv:2306.07299, 2023 - arxiv.org
The Quark-Gluon Plasma (QGP) is a unique phase of nuclear matter, theorized to have filled
the Universe shortly after the Big Bang. A critical challenge in studying the QGP is that, to …

Hierarchical shrinkage Gaussian processes: applications to computer code emulation and dynamical system recovery

T Tang, S Mak, D Dunson - SIAM/ASA Journal on Uncertainty Quantification, 2024 - SIAM
In many areas of science and engineering, computer simulations are widely used as proxies
for physical experiments, which can be infeasible or unethical. Such simulations are often …

Gaussian process subspace prediction for model reduction

R Zhang, S Mak, D Dunson - SIAM Journal on Scientific Computing, 2022 - SIAM
Subspace-valued functions arise in a wide range of problems, including parametric reduced
order modeling (PROM), parameter reduction, and subspace tracking. In PROM, each …

Conglomerate multi-fidelity Gaussian process modeling, with application to heavy-ion collisions

Y Ji, HS Yuchi, D Soeder, JF Paquet, SA Bass… - SIAM/ASA Journal on …, 2024 - SIAM
In an era where scientific experimentation is often costly, multi-fidelity emulation provides a
powerful tool for predictive scientific computing. While there has been notable work on multi …

A hierarchical expected improvement method for Bayesian optimization

Z Chen, S Mak, CF Wu - arXiv preprint arXiv:1911.07285, 2019 - arxiv.org
The Expected Improvement (EI) method, proposed by Jones et al.(1998), is a widely-used
Bayesian optimization method, which makes use of a fitted Gaussian process model for …