Loss-driven sampling within hard-to-learn areas for simulation-based neural network training - Archive ouverte HAL
Conference Papers Year : 2023

Loss-driven sampling within hard-to-learn areas for simulation-based neural network training

Sofya Dymchenko
Bruno Raffin

Abstract

This paper focuses on active learning methods for training neural networks from synthetic input samples that can be generated on-demand. This includes Physics Informed Neural Networks (PINNs), simulation-based inference, deep surrogates and deep reinforcement learning. An adaptive process observes the training progress and steers the data generation with the goal of speeding up and increasing the quality of training. We propose a novel adaptive sampling method that concentrates samples close to the areas showing high loss values. Compared to the state-ofthe-art R3 sampling our algorithm converges to a validation loss of 0.5 in 6000 iterations, while it takes 25000 iterations to reach a loss of 0.7 for the R3 algorithm when training a PINN with the Allen Cahn equation.
Fichier principal
Vignette du fichier
NeurIPS_camera_ready.pdf (427.84 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04305233 , version 1 (24-11-2023)

Licence

Identifiers

  • HAL Id : hal-04305233 , version 1

Cite

Sofya Dymchenko, Bruno Raffin. Loss-driven sampling within hard-to-learn areas for simulation-based neural network training. MLPS 2023 - Machine Learning and the Physical Sciences Workshop at NeurIPS 2023 - 37th conference on Neural Information Processing Systems, Dec 2023, New Orleans, United States. pp.1-5. ⟨hal-04305233⟩
178 View
208 Download

Share

More