A geometrically derived optimizer that rescales gradients by 1/(1+ξ||∇L||²) is competitive with AdamW and Muon, with one RMS-based variant showing slight average improvement.
Visualizing the loss landscape of neural nets,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
The Optimiser Hidden in Plain Sight: Training with the Loss Landscape's Induced Metric
A geometrically derived optimizer that rescales gradients by 1/(1+ξ||∇L||²) is competitive with AdamW and Muon, with one RMS-based variant showing slight average improvement.