Research map / Optimization / Nonconvex and smooth optimization
Boosting First-Order Methods by Shifting Objective: New Schemes with Faster Worst-Case Rates
Kaiwen Zhou, Anthony Man-Cho So, James Cheng
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 |
|---|---|---|---|
| minimize shifted objective in BS-Point-SAGA instead of Point-SAGA | Point-SAGA · worst-case rate factor, large kappa | expected iteration complexity: almost 2 times lower | Improved |
| use G-TM (shifted objective) instead of Triple Momentum | Triple Momentum (TM) · quadratic simulation | convergence rate | No clear effect |
| use G-TM instead of TM in the first iterations | Triple Momentum (TM) · l2-logistic regression on LIBSVM datasets | function value suboptimality: first 30 iterations | Improved |
| use G-TM or TM instead of NAG | Nesterov accelerated gradient · l2-logistic regression on LIBSVM datasets | function value suboptimality | Improved |
| use BS-SVRG instead of SAGA and Katyusha | stochastic first-order methods · a9a, l2-logistic regression | function value suboptimality vs data passes | Mixed |
| output x-tilde instead of z for BS-SVRG (ablation) | BS-SVRG · a9a | early-stage performance | Improved |
| use numerical parameter choice (BS-SVRG-N) instead of analytic choice (ablation) | BS-SVRG · l2-logistic regression | function value suboptimality | No clear effect |
Closest papers by meaning
- Faster Accelerated First-order Methods for Convex Optimization with Strongly Convex Function Constraints (NeurIPS 2024 · similarity 0.89)
- A Universal Transfer Theorem for Convex Optimization Algorithms Using Inexact First-order Oracles (ICML 2024 · similarity 0.88)
- A Study of First-Order Methods with a Deterministic Relative-Error Gradient Oracle (ICML 2024 · similarity 0.87)
- Stochastic Zeroth-Order Optimization under Strongly Convexity and Lipschitz Hessian: Minimax Sample Complexity (NeurIPS 2024 · similarity 0.87)
- Optimizing $(L_0, L_1)$-Smooth Functions by Gradient Methods (ICLR 2025 · similarity 0.87)
- Adaptive and Optimal Second-order Optimistic Methods for Minimax Optimization (NeurIPS 2024 · similarity 0.87)
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.
