Grokking through the Lens of Minimum-Norm Interpolation
cs.LG, math.ST, stat.ML, stat.TH
Submitted: 2026-09-29
Updated: 2026-09-29
Terminology
Sources
- Diagonal Linear Networks and the Lasso Regularization Path
- Minimum Norm Interpolation via the Local Theory of Banach Spaces: The Role of $2$-Uniform Convexity
- Minimum Norm Interpolation via The Local Theory of Banach Spaces: The Role of Gaussianity
- Minimum $\ell_{1}$-norm interpolators: Precise asymptotics and multiple descent
- Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets
- A framework to characterize performance of LASSO algorithms
- The Gaussian min-max theorem in the Presence of Convexity
- Predicting Grokking Long Before it Happens: A look into the loss landscape of models which grok
- Optimal Implicit Bias in Linear Regression
- Explaining grokking through circuit efficiency
- Approaching Deep Learning through the Spectral Dynamics of Weights
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks