Research map / Optimization / Nonconvex and smooth optimization
Utilizing Second Order Information in Minibatch Stochastic Variance Reduced Proximal Iterations
Jialei Wang, Tong Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 MB-SVRP instead of SVRG with weak regularization lambda = 10^-2/n | ridge and logistic regression · codrna, covtype, svmguide3, synthetic | objective suboptimality vs effective passes | Improved |
| use MB-SVRP with strong regularization lambda = 1/n | ridge and logistic regression · codrna, covtype, svmguide3, synthetic | objective suboptimality vs effective passes | No clear effect |
| use minibatch SAGA instead of SAGA | SAGA · smooth regression and classification | convergence speed | Hurt |
| use MB-SVRP-II instead of Option I (ablation) | MB-SVRP · smooth optimization | convergence | No clear effect |
| use prox MB-SVRP with elastic-net regularization at small lambda | prox-SVRG, prox-SAGA · elastic-net logistic and linear regression | objective suboptimality vs effective passes | Improved |
| vary minibatch size b over 16, 32, 64, 128 (ablation) | MB-SVRP · lambda = 10^-2/n | convergence: 16, 32, 64, 128 | No clear effect |
| use L-BFGS instead of variance-reduced methods when lambda is large | L-BFGS · smooth optimization | convergence | Hurt |
Closest papers by meaning
- Second-order Information Promotes Mini-Batch Robustness in Variance-Reduced Gradients (JMLR 2025 · similarity 0.91)
- An Even More Optimal Stochastic Optimization Algorithm: Minibatching and Interpolation Learning (NeurIPS 2021 · similarity 0.91)
- Hybrid Stochastic-Deterministic Minibatch Proximal Gradient: Less-Than-Single-Pass Optimization with Nearly Optimal Generalization (ICML 2020 · similarity 0.91)
- Sharper Analysis for Minibatch Stochastic Proximal Point Methods: Stability, Smoothness, and Deviation (JMLR 2023 · similarity 0.91)
- A Hybrid Stochastic-Deterministic Minibatch Proximal Gradient Method for Efficient Optimization and Generalization (TPAMI 2021 · similarity 0.90)
- Doubly Accelerated Stochastic Variance Reduced Dual Averaging Method for Regularized Empirical Risk Minimization (NeurIPS 2017 · similarity 0.90)
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.
