Semi-supervised impedance inversion by Bayesian neural network based on 2-d CNN pre-training
Muyang Ge, Wangxiangming Zheng, Wenlong Wang
Abstract
Seismic impedance inversion can be performed with a semi-supervised learning algorithm. In this abstract, we improve the semi-supervised learning from two aspects. First, by replacing 1-d convolutional neural network (CNN) layers in deep learning structure with 2-d CNN layers and 2-d maxpooling layers, the prediction accuracy is improved. Second, prediction uncertainty can also be estimated by embedding the network into a Bayesian inference framework. Local reparameterization trick is used during forward propagation of the network to reduce sampling cost. Tests with synthetic data validate the feasibility of the proposed strategy.
Topics & Concepts
Computer scienceArtificial intelligenceConvolutional neural networkInferenceArtificial neural networkEmbeddingMachine learningInversion (geology)Deep learningBayesian inferenceBayesian probabilityPattern recognition (psychology)Bayesian networkSupervised learningPaleontologyBiologyStructural basinSeismic Imaging and Inversion TechniquesGeophysical Methods and ApplicationsSeismic Waves and Analysis