natural-gradient

Tag

Cards List
#natural-gradient

Fisher8: Stabilizing Neural Heteroscedastic Regression via Output-Layer Fisher Geometry

arXiv cs.LG ↗ · 2026-08-12 Cached

This paper introduces Fisher8, an output-layer gradient correction that uses Fisher geometry instead of Euclidean geometry to stabilize neural heteroscedastic regression, improving uncertainty calibration and likelihood-error tradeoffs.

0 favorites 0 likes
#natural-gradient

Between Gradient and Natural Gradient: A Continuum of LoRA Initializations

arXiv cs.LG ↗ · 2026-07-30 Cached

This paper proposes Unified LoRA (ULoRA), a two-parameter family of preconditioned gradient initializations for low-rank adaptation, showing that existing LoRA initialization methods are points on a continuum. The authors demonstrate that a tuned ULoRA matches or exceeds full fine-tuning on GLUE tasks with RoBERTa-base and is competitive on GSM8K with LLaMA 2-7B, and introduce ULoRA-Auto for zero-search deployment.

0 favorites 0 likes
#natural-gradient

Energy Manifold Natural Gradient Descent: Riemannian Optimization for Neural PDE Solvers

arXiv cs.LG ↗ · 2026-07-27 Cached

Introduces Energy Manifold Natural Gradient Descent (EMNGD), a manifold optimization framework for neural PDE solvers that aligns parameter updates with function-space energy curvature while respecting parameter constraints. Theoretical guarantees and empirical results show improved accuracy and convergence.

0 favorites 0 likes
#natural-gradient

Geometric Self-Distillation for Reasoning Generalization

arXiv cs.LG ↗ · 2026-07-09 Cached

This paper introduces GeoSD, a geometric self-distillation objective that uses Hellinger loss and a proximal Fisher-Rao distance term to counter drift in on-policy self-distillation, improving out-of-distribution reasoning accuracy by 5.7–8.6 points across model scales.

0 favorites 0 likes
← Back to home

Submit Feedback