Continual Classification Beyond the Kernel Regime: Harnessing the Sparsity Bias of L2 Regularization in Diagonal Linear Networks
cs.LG
Submitted: 2026-11-04
Updated: 2026-11-04
Terminology
Sources
- Existence of solutions to the nonlinear equations characterizing the precise error of M-estimators
- Generalisation Guarantees for Continual Learning with Orthogonal Gradient Descent
- RAPTOR: Ridge-Adaptive Logistic Probes
- Score-Preserving Continual Lasso under Task Shift
- Optimal Explicit and Implicit Regularization Strengths in High-Dimensional Continual Linear Regression
- In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning
- High-dimensional Asymptotics of Generalization Performance in Continual Ridge Regression
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis