A hybrid LS/GD training scheme for DeepONets, using a factored least-squares solve for the last layer, converges in far fewer work units than Adam-only training on advection, diffusion-reaction, and Poisson benchmarks.
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
CONDITIONAL 1representative citing papers
citing papers explorer
-
Hybrid Least Squares/Gradient Descent Methods for DeepONets
A hybrid LS/GD training scheme for DeepONets, using a factored least-squares solve for the last layer, converges in far fewer work units than Adam-only training on advection, diffusion-reaction, and Poisson benchmarks.