Results 91 to 100 of about 5,348 (248)

Dictionary‐based weak‐form training for noise‐robust series hybrid models with multiplicative unknowns

open access: yesAIChE Journal, EarlyView.
ABSTRACT Hybrid modeling combines first‐principles equations with a data‐driven subcomponent. Training for the data‐driven part is sensitive to measurement noise when training targets are constructed using pointwise time derivatives. Beyond differentiation errors, hybrid models involve solving an inverse problem to estimate the data‐driven term, which ...
Hangjun Cho   +4 more
wiley   +1 more source

High-order methods on mixed-element unstructured meshes for aeronautical applications [PDF]

open access: yes, 2012
Higher resolution and reliability are the desiderata for Computational Fluid Dynamics and main drivers for the development, implementation and validation of highorder accurate methods.
Antoniadis, Antonis F.
core  

Gaussian Process Regression–Neural Network Hybrid with Optimized Redundant Coordinates: A New Simple Yet Potent Tool for Scientist's Machine Learning Toolbox

open access: yesAdvanced Intelligent Discovery, EarlyView.
A machine learning method, opt‐GPRNN, is presented that combines the advantages of neural networks and kernel regressions. It is based on additive GPR in optimized redundant coordinates and allows building a representation of the target with a small number of terms while avoiding overfitting when the number of terms is larger than optimal.
Sergei Manzhos, Manabu Ihara
wiley   +1 more source

Convergence of a high-order compact finite difference scheme for a nonlinear Black-Scholes equation [PDF]

open access: yes
A high-order compact finite difference scheme for a fully nonlinear parabolic differential equation is analyzed. The equation arises in the modeling of option prices in financial markets with transaction costs.
Michel Fournié   +2 more
core  

Explaining the Origin of Negative Poisson's Ratio in Amorphous Networks With Machine Learning

open access: yesAdvanced Intelligent Discovery, EarlyView.
This review summarizes how machine learning (ML) breaks the “vicious cycle” in designing auxetic amorphous networks. By transitioning from traditional “black‐box” optimization to an interpretable “AI‐Physics” closed‐loop paradigm, ML is shown to not only discover highly optimized structures—such as all‐convex polygon networks—but also unveil hidden ...
Shengyu Lu, Xiangying Shen
wiley   +1 more source

Validity of the hyperbolic Whitham modulation equations in Sobolev spaces

open access: yes, 2020
It is proved that modulation in time and space of periodic wave trains, of the defocussing nonlinear Schrödinger equation, can be approximated by solutions of the Whitham modulation equations, in the hyperbolic case, on a natural time scale.
Kostianko, Anna   +2 more
core   +1 more source

Machine‐Learning‐Assisted Onset‐Time Determination in Transient Luminescence Thermometry

open access: yesAdvanced Intelligent Discovery, EarlyView.
Artificial neural networks enable autonomous extraction of onset times from transient heating curves in luminescence thermometry. Using Ln3+‐doped upconverting nanoparticles as luminescent thermometers, we combine experimental transients with physically motivated synthetic curves to enhance data diversity and improve generalization.
David J. Sousa   +3 more
wiley   +1 more source

A class of abstract quasi-linear evolution equations of second order

open access: yes, 1999
In this paper we study the abstract quasi-linear evolution equation of second order formula here in a general banach space z. it is well-known that the abstract quasi-linear theory due to kato [10, 11] is widely applicable to quasi-linear partial ...
NAOKI TANAKA, Tanaka, Naoki
core   +1 more source

Regularity for Semilinear Neutral Hyperbolic Equations with Cosine Families

open access: yes, 2020
The purpose of this paper is to obtain the regularity for solutions of semilinear neutral hyperbolic equations with the nonlinear convolution. The principal operator is the infinitesimal generator of a cosine and sine families.
Jin-Mun Jeong, Seong-Ho Cho
core   +1 more source

A Critical Assessment of Bonding Descriptors for Predicting Materials Properties

open access: yesAdvanced Intelligent Discovery, EarlyView.
The impact of new bonding descriptors in machine learning models for predicting material properties is assessed. Improvements are validated using significance tests, and new, intuitive descriptors for screening lattice thermal conductivity and projected force constants are introduced.
Aakash Ashok Naik   +6 more
wiley   +1 more source

Home - About - Disclaimer - Privacy