Overfitting of Spectral Gradient Descent: How Matrix Geometry shapes Generalization and Implicit Bias

arXiv:2609.32270 · stat.ML, cs.LG · Submitted 2026-09-26 · Read on arXiv

stat.ML, cs.LG

Submitted: 2026-09-26

Updated: 2026-09-26

Terminology

• Generalization — 51× in this paper · explained in 13 episodes
• Trace — 5× in this paper · explained in 2 episodes
• Adam — 4× in this paper · explained in 2 episodes
• Alignment — 3× in this paper · explained in 1 episode
• Transformer — 3× in this paper · explained in 2 episodes
• Anisotropy — 2× in this paper · explained in 3 episodes
• Abundance — 1× in this paper · explained in 1 episode
• Additivity — 1× in this paper · explained in 1 episode
• Ambient Dimension — 1× in this paper · explained in 1 episode
• Attention Mechanism — 1× in this paper · explained in 3 episodes
• Benchmark — 1× in this paper · explained in 2 episodes
• Core — 1× in this paper · explained in 2 episodes
• Degeneracy — 1× in this paper · explained in 3 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Regularization — 1× in this paper · explained in 3 episodes

Sources

• Why do partitions occur in Faa di Bruno's chain rule for higher derivatives?

Related papers

• Behavior of prediction performance metrics with rare events
• Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
• A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
• One Permutation Is All You Need: Fast, Deterministic Feature Importance and Model Stress-Testing
• Online Conformal Prediction for Non-Exchangeable Panel Data
• Deep Time-Series Forecasting in 10 Years: A Survey