This paper deals with the design methodology of an Inverse Neural Network (INN) model. The basic idea is to carry out a semi-physical model gathering two types of information: the a priori knowledge of the deterministic rules which govern the studied system and the observation of the actual conduct of this system obtained from experimental data. This hybrid model is elaborated by being inspired by the mechanisms of a neuromimetic network whose structure is constrained by the discrete reverse-time state-space equations. In order to validate the approach, some tests are performed on two dynamic models. The first suggested model is a dynamic system characterized by an unspecified r-order Ordinary Differential Equation (ODE). The second one concerns in particular the mass balance equation for a dispersion phenomenon governed by a Partial Differential Equation (PDE) discretized on a basic mesh. The performances are numerically analyzed in terms of generalization, regularization and training effort.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.neunet.2012.12.003 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Analysis of a class of two-delay fractional differential equation.
Chaos
January 2025
School of Mathematics and Statistics, University of Hyderabad, Hyderabad 500046, India.
The differential equations involving two discrete delays are helpful in modeling two different processes in one model. We provide the stability and bifurcation analysis in the fractional order delay differential equation Dαx(t)=ax(t)+bx(t-τ)-bx(t-2τ) in the ab-plane. Various regions of stability include stable, unstable, single stable region (SSR), and stability switch (SS).
View Article and Find Full Text PDFAttosecond Rescattering of Laser-Assisted Electron-Proton Collision in Coulomb Potential.
J Phys Chem A
January 2025
Department of Chemistry, University of Malaya, 50603 Kuala Lumpur, Malaysia.
This study investigates the motion of an electron in a Coulomb potential driven by an intense linearly polarized XUV laser pulse analyzed using Gordon-Volkov wave functions. The wave function is decomposed into spherical partial waves to model the scattered electron wave packet after the recollision with a proton. This interaction triggers high harmonic generation, producing coherent X-ray pulses with frequencies that are integer multiples of the XUV field.
View Article and Find Full Text PDFPredicting biomass conversion and COD removal in wastewater treatment by phototrophic bacteria with interpretable machine learning.
J Environ Manage
January 2025
School of Energy & Environmental Engineering, Hebei University of Technology, Tianjin, 300401, China. Electronic address:
Photosynthetic bacteria (PSB) excel in wastewater treatment by removing pollutants and generating biomass but are challenging to optimize due to complex operational and environmental interactions. Neural Ordinary Differential Equations, Elastic Net, Stacking, and Categorical Boosting were applied as artificial intelligence methods to predict chemical oxygen demand (COD) removal efficiency, biomass productivity, biomass yield, and energy yield. Among these, the Stacking model demonstrated superior predictive performance across all targets.
View Article and Find Full Text PDFRevisiting secondary model features for describing the shoulder and lag parameters of microbial inactivation and growth models.
Int J Food Microbiol
January 2025
Departamento de Ingeniería de Alimentos y del Equipamiento Agrícola, Instituto de Biotecnología Vegetal, Universidad Politécnica de Cartagena (ETSIA), Paseo Alfonso XIII, 48, 30203 Cartagena, Spain.
The Baranyi and Geeraerd models are two of the most reliable models for the description of, respectively, microbial growth and inactivation. They are defined as a system of differential equations, whose algebraic solution can describe the microbial response during isothermal conditions, especially when combined with suitable secondary models. However, there are still large uncertainties regarding the best functions to use as secondary models for the lag phase duration (λ) and the shoulder length (S).
View Article and Find Full Text PDFPrecision of in vivo pressure gradient estimations using synthetic aperture ultrasound.
Ultrasonics
January 2025
The Center for Fast Ultrasound Imaging, Department of Health Technology. Technical University of Denmark, Ørsteds Plads Building 349, Lyngby, DK-2800, Denmark.
Non-invasive estimation of pressure differences using 2D synthetic aperture ultrasound imaging offers a precise, low-cost, and risk-free diagnostic tool. Unlike invasive techniques, this preserves natural blood flow and avoids the limitations of devices that occupy lumen space. This paper evaluates a previously published estimator, modified to incorporate Singular Value Decomposition (SVD) echo-cancellation, using data from ten healthy volunteers and one patient.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!