Skip to Main content Skip to Navigation
Other publications

A Taylor Based Sampling Scheme for Machine Learning in Computational Physics

Abstract : Machine Learning (ML) is increasingly used to construct surrogate models for physical simulations. We take advantage of the ability to generate data using numerical simulations programs to train ML models better and achieve accuracy gain with no performance cost. We elaborate a new data sampling scheme based on Taylor approximation to reduce the error of a Deep Neural Network (DNN) when learning the solution of an ordinary differential equations (ODE) system.
Complete list of metadata

https://hal.inria.fr/hal-03114984
Contributor : Paul Novello Connect in order to contact the contributor
Submitted on : Thursday, January 28, 2021 - 10:49:31 AM
Last modification on : Friday, April 30, 2021 - 10:05:06 AM

Files

NeurIPS_2019_workshop_final.pd...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03114984, version 2
  • ARXIV : 2101.11105

Citation

Paul Novello, Gaël Poëtte, David Lugato, Pietro Congedo. A Taylor Based Sampling Scheme for Machine Learning in Computational Physics. 2019. ⟨hal-03114984v2⟩

Share

Metrics

Les métriques sont temporairement indisponibles