Multi-Output Deep Ensemble for a 1D Elastic Bar¶

Notebook: MultiOutputDeepEnsemble_ElasticBar1D_Tutorial.ipynb

This tutorial uses MultiOutputDeepEnsembleRegressor on a 1D elastic bar. The model predicts two coupled scientific outputs at once: displacement and stress.

Key ideas: - multi-output regression under parameter variation, - simultaneous uncertainty bands for displacement and stress, - OOD cases with stronger loading and stiffness gradients.

Primary references: - Lakshminarayanan, Pritzel, Blundell (2017), Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles. NeurIPS proceedings - Nix, Weigend (1994), Estimating the Mean and Variance of the Target Probability Distribution. DOI: 10.1109/ICNN.1994.374138