Bridging Deep Convolutional Autoencoders and Ensemble Smoothers for Improved Estimation of Channelized Reservoirs

Data‐Worth Analysis for Heterogeneous Subsurface Structure Identification With a Stochastic Deep Learning Framework

“…In the last few years, more and more efficient and accurate DA‐based methods for subsurface structure identification have been proposed by improving one or more of the modules mentioned above (Jardani et al., 2022; Jo et al., 2021; Kang et al., 2021; Mo et al., 2020; Sebacher & Toma, 2022). Within the context of reducing first arrival time uncertainty, Rizzo and de Barros (2019) proposed an iterative sampling method that relies on a graph theory‐based connectivity metric to capture preferential flow paths.…”