Research map / Optimization / Nonconvex and smooth optimization
Variance Reduction via Primal-Dual Accelerated Dual Averaging for Nonsmooth Convex Finite-Sums
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 VRPDA² instead of the prior smoothing-based finite-sum approach | Structured nonsmooth finite-sum optimization · General convex primal-dual gap | overall complexity: O(nd log min{1/ε, n} + √n d/ε) -> O(nd log min{1/ε, n} + d/ε) | Helped |
| Use VRPDA² instead of the prior smoothing-based finite-sum approach | Structured nonsmooth finite-sum optimization · Strongly convex primal-dual gap | overall complexity: O(nd log min{1/ε, n} + √n d/√ε) -> O(nd log min{1/ε, n} + d/√(σε)) | Helped |
| Use the proposed coordinate primal-dual updates | Elastic net-regularized SVM · a9a and MNIST | primal function value gap | Helped |
| Use the average iterate rather than the last iterate | VRPDA² · Elastic net-regularized SVM | primal function value gap | Helped |
| Use the deterministic full primal-dual initialization before randomized updates | VRPDA² · Elastic net-regularized SVM | primal function value gap | No clear effect |
| Use VRPDA² instead of SPDHG or PURE_CD | Elastic net-regularized SVM · a9a and MNIST | number of nonzeros | Helped |
Closest papers by meaning
- Variance Reduction via Accelerated Dual Averaging for Finite-Sum Optimization (NeurIPS 2020 · similarity 0.94)
- Doubly Accelerated Stochastic Variance Reduced Dual Averaging Method for Regularized Empirical Risk Minimization (NeurIPS 2017 · similarity 0.90)
- Kill a Bird with Two Stones: Closing the Convergence Gaps in Non-Strongly Convex Optimization by Directly Accelerated SVRG with Double Compensation and Snapshots (ICML 2022 · similarity 0.89)
- RandProx: Primal-Dual Optimization Algorithms with Randomized Proximal Updates (ICLR 2023 · similarity 0.89)
- Double Variance Reduction: A Smoothing Trick for Composite Optimization Problems without First-Order Gradient (ICML 2024 · similarity 0.87)
- Drago: Primal-Dual Coupled Variance Reduction for Faster Distributionally Robust Optimization (NeurIPS 2024 · similarity 0.86)
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.