A Comparison Between Stacked Auto-Encoder and Deep Belief Network in River Run-Off Prediction

A Systematic Review of Deep Learning Applications in Streamflow Data Augmentation and Forecasting

“…Qian et al (2020) improved their loss function for their sparse autoencoder model with the inverse method of simulated annealing (ESA) for the runoff of the Kenswat Station in the Manas River Basin in northern Xinjiang, China. In DBN-based studies from 2020,Kinh et al (2020) for the Srepok River in the Central Highlands of Vietnam, compared a SAE model to a DBN model that showed similar performance Zhan et al (2020). incorporated variational inference into the BNN model and proposed a variational Bayesian neural network (VBNN) model for ensemble daily mean flood forecasting.…”