A trainable-by-parts variational autoencoder with a neural network mapper claims to beat FNO and DeepONet on groundwater flow problems in both accuracy and training efficiency.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
VAE-DNN: Energy-Efficient Trainable-by-Parts Surrogate Model For Parametric Partial Differential Equations
A trainable-by-parts variational autoencoder with a neural network mapper claims to beat FNO and DeepONet on groundwater flow problems in both accuracy and training efficiency.