Gradient Descent on Logistic Regression with Non-Separable Data and Large Step Sizes
cs.LG, math.OC
Submitted: 2024-06-07
Updated: 2026-09-02
Terminology
Sources
- Acceleration by Stepsize Hedging I: Multi-Step Descent and the Silver Stepsize Schedule
- Layer Normalization
- Self-concordant analysis for logistic regression
- Squareplus: A Softplus-Like Algebraic Rectifier
- Beyond the Edge of Stability via Two-step Gradient Updates
- Handbook of Convergence Theorems for (Stochastic) Gradient Methods
- Accelerated Gradient Descent via Long Steps
- Directional Smoothness and Gradient Methods: Convergence and Adaptivity
- Good regularity creates large learning rate implicit biases: edge of stability, balancing, and catapult
- Quadratic models for understanding catapult dynamics of neural networks
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