Blockwise Stabilized Adaptive Cubic Regularization with Subsolvers via Recurrence
cs.LG, cs.NA, math.NA
Submitted: 2026-08-22
Updated: 2026-10-14
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Random Subspace Cubic-Regularization Methods, with Applications to Low-Rank Functions
- Extended-Krylov-subspace methods for trust-region and norm-regularization subproblems
- Block cubic Newton with greedy selection
- Symmetric Rank-One Quasi-Newton Methods for Deep Learning Using Cubic Regularization
- Chebyshev-Exact Acceleration under Hessian Variation, I: Sine-Jacobi Method
- Scalable Second-Order Optimization Algorithms for Minimizing Low-rank Functions
- On Global Rates for Regularization Methods based on Secant Derivative Approximations
- A General Recipe for Parameter-Free Nonconvex Optimization via Higher-Order Regularization
- Parameter-Free Cubic-Regularized Newton Method: Sharp Complexity and Generalized Smoothness
- Adaptive Regularized Newton Method with Inexact Hessian
- Cubic Regularized Newton Method with Variance Reduction for Finite-sum Non-convex Problems
- First- and Second-Order Stochastic Adaptive Regularization with Cubics: High Probability Iteration and Sample Complexity
- Faster Newton Methods for Convex and Nonconvex Optimization in Gradient Complexity
- Adaptive Newton-CG methods with global and local analysis for unconstrained optimization with H\"older continuous Hessian
- A MINRES-based Linesearch Algorithm for Nonconvex Optimization with Non-positive Curvature Detection
- Towards Practical Second-Order Optimizers in Deep Learning: Insights from Fisher Information Analysis
- A simple and practical adaptive trust-region method
- Adaptive Regularization within Trust Region Methods for Stochastic Nonconvex Optimization
- A scalable sequential adaptive cubic regularization algorithm for optimization with general equality constraints
- A Fully Parameter-Free Second-Order Algorithm for Convex-Concave Minimax Problems
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: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks