mlsanityData to 7 Oct 2026

Training Variation of Physically-Informed Deep Learning Models

arXiv 2510.03416 ↗First posted 3 Oct 2025cs.LG cond-mat.mtrl-scicited by 0

Not on the map. Why?

Abstract (arXiv)

A successful deep learning network is highly dependent not only on the training dataset, but the training algorithm used to condition the network for a given task. The loss function, dataset, and tuning of hyperparameters all play an essential role in training a network, yet there is not much discussion on the reliability or reproducibility of a training algorithm. With the rise in popularity of physics-informed loss functions, this raises the question of how reliable one's loss function is in conditioning a network to enforce a particular boundary condition. Reporting the model variation is needed to assess a loss function's ability to consistently train a network to obey a given boundary condition, and provides a fairer comparison among different methods. In this work, a Pix2Pix network predicting the stress fields of high elastic contrast composites is used as a case study. Several different loss functions enforcing stress equilibrium are implemented, with each displaying different levels of variation in convergence, accuracy, and enforcing stress equilibrium across many training sessions. Suggested practices in reporting model variation are also shared.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 13 of its 35 references that are in the map. First posted on arXiv 3 Oct 2025.

Builds on

13 of 35 references.

References · 13

  1. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  2. 2021Characterizing possible failure modes in physics-informed neural networkscited by 632
  3. 2022Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's nextcited by 603
  4. 2020Improving Reproducibility in Machine Learning Research (A Report from the NeurIPS 2019 Reproducibility Program)cited by 318
  5. 2021Accounting for Variance in Machine Learning Benchmarkscited by 195
  6. 2019The Convergence Rate of Neural Networks for Learned Functions of Different Frequenciescited by 134
  7. 2022When Physics Meets Machine Learning: A Survey of Physics-Informed Machine Learningcited by 64
  8. 2020Deep learning for synthetic microstructure generation in a materials-by-design framework for heterogeneous energetic materialscited by 18
  9. 2022Improving the repeatability of deep learning models with Monte Carlo dropoutcited by 11
  10. 2022Three-dimensional microstructure generation using generative adversarial neural networks in the context of continuum micromechanicscited by 11
  11. 2021Uncertainty Prediction for Machine Learning Models of Material Propertiescited by 10
  12. 2018Study and Observation of the Variation of Accuracies of KNN, SVM, LMNN, ENN Algorithms on Eleven Different Datasets from UCI Machine Learning Repositorycited by 4
1 more
  1. 2021On the reproducibility of fully convolutional neural networks for modeling time-space evolving physical systemscited by 3

Built on by

Not cited yet.