Full text loading...
-
A Monte Carlo Dropout Based Deep Learning Model to Quantify Uncertainty in Facies Classification
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, EAGE Workshop on Quantitative Geoscience as a Catalyst in a Carbon Neutral World, May 2022, Volume 2022, p.1 - 3
Abstract
Summary
By measuring the parameter uncertainty of the deep neural net. weights and biases by adding an additional Monte Carlo Dropout layer, the proposed Monte Carlo Dropout based Deep Neural Net. quantifies the uncertainty in the process of facies classification using seismic data and outputs a better classification than the traditional Deep Learning model.
© EAGE Publications BV