Results 211 to 220 of about 7,749,372 (301)

Electrospun Porous Materials for the Next‐Generation Proton Exchange Membrane Fuel Cells: Progress, Challenges, and AI‐Optimized Perspectives

Carbon Energy, EarlyView.
Incorporating AI to optimize solution, process, and environment enables efficient prediction of structural properties and performance, allowing tailoring of GDL, CL, and PEM fabrication to guide PEMFC toward 9.0 kW L‒1 by 2040. ABSTRACT The performance and commercial viability of proton exchange membrane fuel cells (PEMFCs) are constrained by key ...
Shangfeng Jiang, Panfeng Wang, You Gong, Zeyu Sun, Huatan Chen, Geng Wang, Ting Tao, Chuang Zhai, Xi Wang, Ruixin Chen, Jiaxin Jiang, Bingrong Zhang   +11 more
wiley   +1 more source

A neuro-inspired computational framework for AGI: predictive coding, active inference and free energy minimisation. [PDF]

Front Hum Neurosci
Berndt LCS, Shaw AD.
europepmc   +1 more source

Laminar Burning Velocity of Ammonia‐Hydrogen Mixtures Predicted by Machine Learning

Chemie Ingenieur Technik, EarlyView.
Ammonia–hydrogen blends aid defossilization, but slow combustion kinetics require complex modeling. Although chemical kinetic modeling is computationally expensive, machine learning surrogates offer fast predictions. This study trained GPR, DT, and ANN models on 1445 experimental points.
Georg Klepp, Aditya Sharma, Helene Dörksen   +2 more
wiley   +1 more source

Correction to: 'Imputation of body mass' (2021) by Turvey et al. [PDF]

Proc Biol Sci
Turvey ST, Duncan C, Upham N, Harrison XA, Dávalos LM.   +4 more
europepmc   +1 more source

Harnessing machine learning and optimization for informed chemical engineering decisions: A styrene reactor analysis

The Canadian Journal of Chemical Engineering, EarlyView.
This study shows that integrating multiple machine learning models with optimization and decision‐making improves chemical process design, and that a consensus‐based strategy across models provides more robust and reliable operating recommendations than any single model, especially under limited or noisy data conditions.
Farough Agin, Clémence Fauteux‐Lefebvre, Jules Thibault   +2 more
wiley   +1 more source

Small dense low-density lipoprotein cholesterol and risk of cardiovascular events in general adults: a systematic review and meta-analysis of prospective studies. [PDF]

BMJ Open
Duc TQ, Nguyen Di K, Dang Khoa D.
europepmc   +1 more source

Hierarchical Bayesian Modelling of Visual Attention

Proceedings of the 9th International Conference on Computer Vision Theory and Applications, 2014
openaire   +2 more sources

Vine copula knockoffs for variable selection in gene expression studies

Canadian Journal of Statistics, EarlyView.
Abstract Identifying clinical and genetic markers is essential for stratifying cancer patients by survival outcomes and guiding personalized treatment strategies. However, gene expression studies often involve high‐dimensional predictors with mixed data types and complex dependence, which complicates reliable variable selection.
José Ulises Márquez Urbina, Alejandro Román Vásquez, Graciela González Farías, Gabriel Escarela   +3 more
wiley   +1 more source

Data-driven exploration of occupations and lung cancer risks using regularized regression models. [PDF]

Int J Epidemiol
Ge C, Nie J, Portengen L, Olsson A, Schüz J, Ahrens W, Barul C, Behrens T, Brüning T, Caporaso NE, Consonni D, Demers PA, Fabianova E, Fernández-Tardón G, Field JK, Forastiere F, Foretova L, Gustavsson P, Janout V, Jöckel KH, Karrasch S, Kendzia B, Kromhout H, Landi MT, Lissowska J, Mates D, McLaughlin JR, Merletti F, Migliore E, Pándics T, Radoi L, Richiardi L, Schejbalova M, Siemiatycki J, Świątkowska B, Wichmann HE, Zaridze D, Straif K, Peters S, Vermeulen R.   +39 more
europepmc   +1 more source

Lasso for hierarchical polynomial models

Canadian Journal of Statistics, EarlyView.
Abstract The divisibility conditions implicit in a polynomial hierarchy suggest parameter constraints in regression. With this idea, we establish strong and weak hierarchies for both the lasso and relaxed lasso. Our proposal extends prior work on hierarchical lasso, which was mainly concerned with models of degree 2.
H. Maruri‐Aguilar, S. Lunagómez
wiley   +1 more source