Code for the graph-neural-network study (Section 5 and Appendix 9) of the article "Modelling magnetic material properties with uncertainty-aware neural networks" (Computational Materials Science 276, 115073, 2027, https://doi.org/10.1016/j.commatsci.2026.115073). A graph neural network predicts the coercivity μ₀H_c of Nd₂Fe₁₄B microstructures and provides aleatoric and epistemic uncertainty estimates (Gaussian negative log-likelihood loss and Monte Carlo dropout); confidence curves assess the quality of the uncertainty estimates. The script reproduce_fig18.py reproduces the results and Fig. 18 of the paper; it is a copy of the corresponding s
Moustafa, H., Wager, C., Kovacs, A., Ali, Q., Oezelt, H., Schrefl, T.
The reproducible core of a study of Bayesian tensor-basis neural network corrections for RANS turbulence closures, with epistemic and aleatoric uncertainty propagated through the CFD solver. Contains the training dataset for six separated flows (periodic hill, curved backward-facing step, three parameterized hills and the NASA wall-mounted hump), the code that trains the scalar and anisotropy correction models, the four trained checkpoints, the OpenFOAM cases with meshes, and the published propagated statistics for comparison against a reproduction. Has no private dependencies: everything except the CFD runs from a clone and the supplied cond
Buchanan, Tyler, Eidi, Ali, Dwight, Richard P.
This repository provides code and workflows to train, evaluate, and analyze an evidential deep neural network (EDNN) emulator for CLM5 (Community Land Model v5) leaf area index (LAI) using a perturbed-parameter ensemble (PPE). The EDNN maps 32 CLM5 parameters to seasonal LAI anomalies and outputs a Normal–Inverse–Gamma predictive distribution, providing the mean and decomposed aleatoric and epistemic uncertainties. The workflow supports seasonal evaluation and FAST (Fourier Amplitude Sensitivity Test) analyses to quantify main and interaction effects across seasons. What's included Data preprocessing for CLM5 PPE outputs and feature engineeri
Kachinga Silwimba
DLLib DLLib is a modular toolkit for image-based deep learning that pairs baseline models with supervision, uncertainty, anomaly detection, and ensemble mechanisms to improve robustness, explainability, and reliability. Highlights - Uncertainty Inference : Aleatoric- and parallel-based transformations to produce diverse, semantically consistent predictions. - Supervision & Explainability : Lightweight checks to surface potential issues during model execution. - Anomaly Detection (VAEs) : Input/output/activation-level outlier detection against the training distribution. - Ensembling : Combines model predictions with supervisor signals for stro
Pujol, Roger, Brando, Axel, Mezzetti, Enrico, Cazorla, Francisco J., Abella, Jaume
deep Learning.py contains the core algorithmic implementation of the proposed Uncertainty-Aware Intelligent Multi-View Clustering (UA-IMVC) framework. The script integrates view-specific deep autoencoders to extract low-dimensional nonlinear latent representations from heterogeneous clinical features, coupled with an evidential deep learning module that estimates Normal-Inverse-Gamma (NIG) parameters to explicitly quantify both aleatoric and epistemic uncertainties. Furthermore, it incorporates a reliability-guided adaptive feature fusion mechanism to aggregate view-specific insights based on their normalized inverse uncertainties, a confiden
norrth-z
A Python software package for designing and developing Bayesian Autoencoder (BAE) which is a class of probabilistic deep learning model. BAE quantifies epistemic and aleatoric uncertainties in their outputs which are not usually available with conventional deep learning models. For more information, please visit my github page here
bangxiangyong
