Results 71 to 80 of about 7,687 (182)

Bayesian OOD detection with aleatoric uncertainty and outlier exposure

, 2021
Typical Bayesian approaches to OOD detection use epistemic uncertainty. Surprisingly from the Bayesian perspective, there are a number of methods that successfully use aleatoric uncertainty to detect OOD points (e.g. Hendryks et al. 2018). In addition, it is difficult to use outlier exposure to improve a Bayesian OOD detection model, as it is not clear
Wang, Xi, Aitchison, Laurence
openaire   +2 more sources

Uncertainty Calibration in Molecular Machine Learning: Comparing Evidential and Ensemble Approaches

Chemistry – A European Journal, Volume 32, Issue 15, 22 April 2026.
Raw uncertainty estimates from deep evidential regression and deep ensembles are systematically miscalibrated. Post hoc calibration aligns predicted uncertainty with true errors, improving reliability and enabling efficient active learning and reducing computational cost while preserving predictive accuracy.
Bidhan Chandra Garain, Max Pinheiro Jr., Matheus de Oliveira Bispo, Mario Barbatti   +3 more
wiley   +1 more source

Machine Learning for Designing Perovskites and Perovskite‐Inspired Solar Materials: Emerging Opportunities and Challenges

Advanced Science, Volume 13, Issue 23, 23 April 2026.
This review offers a comprehensive comparison between perovskites and perovskite‐inspired materials (PIMs), focusing on their crystal structures, electronic properties, and chemical compositions. It evaluates the applicability of machine learning (ML) descriptors and models across both material classes.
Yangfan Zhang, Yiming Xia, Ali Shakiba, Hongrui Zhang, Xiaojing Hao, Priyank V. Kumar, Mahesh P. Suryawanshi   +6 more
wiley   +1 more source

An uncertainty-aware domain adaptive semantic segmentation framework

Autonomous Intelligent Systems
Semantic segmentation is significant to realize the scene understanding of autonomous driving. Due to the lack of annotated real-world data, the technology of domain adaptation is applied so that the model is trained on the synthetic data and inferred on
Huilin Yin, Pengyu Wang, Boyu Liu, Jun Yan   +3 more
doaj   +1 more source

Informing synthetic passive microwave predictions through Bayesian deep learning with uncertainty decomposition

Environmental Data Science
Space-borne passive microwave (PMW) data provide rich information on atmospheric state, including cloud structure and underlying surface properties. However, PMW data are sparse and limited due to low Earth orbit collection, resulting in coarse Earth ...
Pedro Ortiz, Eleanor Casas, Marko Orescanin, Scott W. Powell, Veljko Petkovic   +4 more
doaj   +1 more source

Epistemic Uncertainty Analysis and Robust Optimization of a Second‐Generation Solvent‐Based Post‐Combustion Carbon Capture Process

Engineering Reports, Volume 8, Issue 4, April 2026.
Parametric uncertainty in the PZ/AFS post‐combustion carbon‐capture process model compromises the reliability of process designs extracted via traditional, deterministic optimization methods. We identify the twelve parameters exhibiting highest impact on capture and costs, and mitigate the design risk stemming from those via formulating and solving a ...
Ilayda Akkor, Shachit S. Iyer, John Dowdle, Le Wang, Chrysanthos E. Gounaris   +4 more
wiley   +1 more source

Spatio‐Temporal Modelling of Extreme Induced Seismicity in the Presence of An Evolving Measurement Network

Environmetrics, Volume 37, Issue 3, April 2026.
ABSTRACT Earthquakes induced by injecting or extracting gas from underground reservoirs can pose a significant hazard to surrounding infrastructure and populations. Safeguarding against future seismic hazards requires accurate models for the upper tail of the earthquake magnitude distribution that are able to represent various intervention strategies ...
Conor Murphy, Jonathan A. Tawn, Zak Varty, Ross Towe, Peter M. Atkinson   +4 more
wiley   +1 more source

Integration of Machine Learning With PBPK and QSAR Modeling Approaches to Facilitate Drug Discovery and Development

CPT: Pharmacometrics &Systems Pharmacology, Volume 15, Issue 4, April 2026.
ABSTRACT This review examines the application of machine learning (ML) in physiologically based pharmacokinetic (PBPK) modeling through improved prediction of input parameters, particularly via quantitative structure–activity relationship (QSAR) models, for absorption, distribution, metabolism, and excretion (ADME) properties across drug development ...
Xinyue Chen, Zhoumeng Lin
wiley   +1 more source

Deep Modeling of Non-Gaussian Aleatoric Uncertainty

IEEE Robotics and Automation Letters
Deep learning offers promising new ways to accurately model aleatoric uncertainty in robotic state estimation systems, particularly when the uncertainty distributions do not conform to traditional assumptions of being fixed and Gaussian. In this study, we formulate and evaluate three fundamental deep learning approaches for conditional probability ...
Aastha Acharya, Caleb Lee, Marissa D'Alonzo, Jared Shamwell, Nisar R. Ahmed, Rebecca Russell   +5 more
openaire   +2 more sources

Deformation and Earthquake Potential on the North America–Caribbean–Cocos Plate Boundary System in Guatemala

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 4, April 2026.
Abstract In Guatemala, the Cocos, North American, and Caribbean plates interact to create a region of high seismic risk. Previous analyses of crustal faults in the country have been overly simplified, creating discrepancies between geologic and geodetic slip rate models.
Jeremy Maurer, Andreas Eckert, Qiaoqi Sun, Jonathan Obrist‐Farner   +3 more
wiley   +1 more source