TGLF-WINN matches standard NN surrogate accuracy for TGLF using only 25% of the training data via physics-guided regularization and active learning, delivering 45x speedup in flux-matching workflows.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.plasm-ph 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
TGLF-WINN: Data-Efficient Deep Learning Surrogate for Turbulent Transport Modeling in Fusion
TGLF-WINN matches standard NN surrogate accuracy for TGLF using only 25% of the training data via physics-guided regularization and active learning, delivering 45x speedup in flux-matching workflows.