Results 71 to 80 of about 3,934 (193)

Reliable Intelligent Diagnostics With Uncertainty Quantification for Mechanical System Condition Assessment

Quality and Reliability Engineering International, Volume 42, Issue 6, Page 2896-2916, October 2026.
ABSTRACT Health condition assessment of mechanical systems is essential to ensure their safe and reliable operation. Although AI‐driven diagnostic techniques have advanced significantly with the development of deep learning, the reliability of such techniques is often compromised due to the lack of uncertainty quantification (UQ)—particularly the ...
Guangjun Jiang, Haotian Ouyang, Zifei Xu
wiley   +1 more source

Effects of aleatoric and epistemic errors in reference data on the learnability and quality of NN-based potential energy surfaces

Artificial Intelligence Chemistry
The effect of noise in the input data for learning potential energy surfaces (PESs) based on neural networks for chemical applications is assessed.
Sugata Goswami, Silvan Käser, Raymond J. Bemish, Markus Meuwly   +3 more
doaj   +1 more source

Neural Surrogate HMC: On Using Neural Likelihoods for Hamiltonian Monte Carlo in Simulation‐Based Inference

Journal of Geophysical Research: Machine Learning and Computation, Volume 3, Issue 5, October 2026.
Abstract Bayesian inference methods such as Markov Chain Monte Carlo (MCMC) typically require repeated computations of the likelihood function, but in some scenarios, this is infeasible and alternative methods are needed. Simulation‐based inference methods address this problem by using machine learning to amortize computations.
L. M. Wolniewicz, P. Sadowski, C. Corti
wiley   +1 more source

Disentangling Aleatoric and Epistemic Uncertainty in Physics‐Informed Neural Networks: Application to Insulation Material Degradation Prognostics

Advanced Intelligent Systems
Physics‐Informed Neural Networks (PINNs) provide a framework for integrating physical laws with data. However, their application to Prognostics and Health Management (PHM) remains constrained by the limited uncertainty quantification (UQ) capabilities ...
Ibai Ramirez, Jokin Alcibar, Joel Pino, Mikel Sanz, Jose I. Aizpurua   +4 more
doaj   +1 more source

Site‐Specific Post‐Processing of Spectral Wave Forecast by Learning From Buoy Measurements

Journal of Geophysical Research: Machine Learning and Computation, Volume 3, Issue 5, October 2026.
Abstract Operational wave prediction systems provide critical information across a range of marine industries and for public safety. Yet operational numerical wave forecasts can exhibit systematic biases, such as in wave height and arrival time of swell events.
Arthur Filoche, Jeff Hansen, Kevin Vinsen, Travis Dawson   +3 more
wiley   +1 more source

How Much Physics Should a Neural Network Know? Thermodynamics‐Informed Neural Networks for Rock Constitutive Modeling With Epistemic Uncertainty

Journal of Geophysical Research: Machine Learning and Computation, Volume 3, Issue 5, October 2026.
Abstract Machine learning offers a flexible route to constitutive modeling, with two emergent questions for geoscience applications: what level of thermodynamic constraint should be embedded in the network architecture, and how should predictive uncertainty be quantified when extrapolating from laboratory to field conditions? We address both challenges
Kangan Li, Tushar Mittal, Benjamin K. Holtzman, Seth Saltiel   +3 more
wiley   +1 more source

Improved multi-objective decision-making in manufacturing processes through uncertainty quantification and robust pareto front modelling

Scientific Reports
Manufacturing processes often exhibit complex relationships between input parameters and output responses, posing challenges for optimization and decision-making.
Arne De Temmerman, Mathias Verbeke
doaj   +1 more source

Embracing uncertainty in time: A Bayesian dynamic mixing model for isotopic diet estimation

Methods in Ecology and Evolution, Volume 17, Issue 10, Page 2928-2942, October 2026.
Abstract Isotopic mixing models are powerful tools in ecology for estimating diets. Since their initial formulation, these models have evolved into two distinct branches: Bayesian mixing models and dynamic mixing models. Bayesian mixing models significantly improve uncertainty management but remain static assuming a stable state.
Emilie Cathelin, Sébastien Lefebvre, Carolina Giraldo, Andrew Parnell   +3 more
wiley   +1 more source

Aleatoric and Epistemic Discrimination: Fundamental Limits of Fairness Interventions [PDF]

Machine learning (ML) models can underperform on certain population groups due to choices made during model development and bias inherent in the data. We categorize sources of discrimination in the ML pipeline into two classes: aleatoric discrimination ...
Gao, Rui, Calmon, Flavio P., He, Luxi, Wang, Hao   +3 more
core   +1 more source

Data-driven organic solubility prediction at the limit of aleatoric uncertainty

Nature Communications
Small molecule solubility is a critically important property which affects the efficiency, environmental impact, and phase behavior of synthetic processes.
Lucas Attia, Jackson W. Burns, Patrick S. Doyle, William H. Green   +3 more
doaj   +1 more source