Under a river-valley model of the loss landscape, the paper shows that valley fluctuations behave like heat, with learning rate as temperature, and derives a 1/t optimal decay schedule.
Statistical mechanics of learning
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
-
Neural Thermodynamic Laws for Large Language Model Training
Under a river-valley model of the loss landscape, the paper shows that valley fluctuations behave like heat, with learning rate as temperature, and derives a 1/t optimal decay schedule.