Research map / Optimization / Nonconvex and smooth optimization
Stochastic Variance-Reduced Cubic Regularized Newton Methods
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use semi-stochastic variance-reduced gradient and Hessian estimates in cubic regularization | SVRC · (ε, √ρε)-approximate local minimum | SO calls: O(n/ε3/2) -> eO(n + n4/5/ε3/2) | Helped |
| Use semi-stochastic variance-reduced gradient and Hessian estimates instead of subsampling | SVRC · (ε, √ρε)-approximate local minimum | SO calls: eO(n/ε3/2 + 1/ε5/2) -> eO(n + n4/5/ε3/2) | Helped |
| Use an inexact cubic-subproblem solver satisfying Condition 5.1 (ablation) | SVRC · (ε, √ρε)-approximate local minimum | SO calls: O(n + ∆F√ρn4/5/ε3/2) | No clear effect |
| Use an inexact cubic-subproblem solver satisfying Condition 5.1 (ablation) | SVRC · (ε, √ρε)-approximate local minimum | CSO calls: O(∆F√ρ/ε3/2) | No clear effect |
| Use SVRC instead of the compared baselines | Non-convex regularized logistic regression · a9a, covtype, ijcnn1 | function value gap | Helped |
| Use SVRC instead of the compared baselines | Nonlinear least squares · a9a, ijcnn1 | function value gap | Helped |
| Use SVRC instead of Adaptive Cubic | Nonlinear least squares · covtype | epoch count at high accuracy | Hurt |
| Use SVRC instead of the compared baselines | Nonlinear least squares · covtype | CPU time | Helped |
| Use SVRC instead of the compared baselines | Robust linear regression · a9a, ijcnn1 | function value gap | Helped |
| Use SVRC instead of Adaptive Cubic | Robust linear regression · covtype | epoch count at high accuracy | Hurt |
| Use SVRC instead of the compared baselines | Robust linear regression · covtype | CPU time | Helped |
Closest papers by meaning
- Stochastic Cubic Regularization for Fast Nonconvex Optimization (NeurIPS 2018 · similarity 0.93)
- Simple Stochastic Gradient Methods for Non-Smooth Non-Convex Regularized Optimization (ICML 2019 · similarity 0.91)
- Stochastic Subspace Cubic Newton Method (ICML 2020 · similarity 0.90)
- Stochastic Newton Proximal Extragradient Method (NeurIPS 2024 · similarity 0.88)
- SERENA: A Unified Stochastic Recursive Variance Reduced Gradient Framework for Riemannian Non-Convex Optimization (ICML 2025 · similarity 0.88)
- Double Variance Reduction: A Smoothing Trick for Composite Optimization Problems without First-Order Gradient (ICML 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.