MLchartDataset catalogue

Back propagation

Term · Oil and gas · MLC-T-OAG-000239

The numerical extrapolation, in seismic imaging, of a recorded wavefield backward in time or depth through a velocity model, as in reverse time migration and in the adjoint step of full-waveform inversion. In machine learning the same name is used for the algorithm that trains neural networks by passing the error gradient backward through the layers, a method also applied to seismic noise attenuation and facies classification.

Table 1. Record
IdentifierMLC-T-OAG-000239
FieldOil and gas
SubjectGeophysics
Record as JSON 
{
  "id": "MLC-T-OAG-000239",
  "term": "Back propagation",
  "field": "Oil and gas",
  "definition": "The numerical extrapolation, in seismic imaging, of a recorded wavefield backward in time or depth through a velocity model, as in reverse time migration and in the adjoint step of full-waveform inversion. In machine learning the same name is used for the algorithm that trains neural networks by passing the error gradient backward through the layers, a method also applied to seismic noise attenuation and facies classification.",
  "subject": "Geophysics",
  "url": "https://mlchart.com/terminology/oil-and-gas/back-propagation/"
}

Record 259 of 4,469 in Oil and gas terminology (MLC-0103). Request the full dataset.