Research map / Optimization / Nonconvex and smooth optimization
Faster Stochastic Alternating Direction Method of Multipliers for Nonconvex Optimization
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 the SPIDER recursive gradient estimator in place of the deterministic full-gradient ADMM update | Nonconvex ADMM · Finite-sum nonconvex optimization | IFO complexity: \mathcal{O}(n\epsilon^{-1}) -> \mathcal{O}(n+n^{\frac{1}{2}}\epsilon^{-1}) | Helped |
| Use the SPIDER recursive gradient estimator instead of the SVRG-ADMM estimator | Nonconvex stochastic ADMM · Finite-sum nonconvex optimization | IFO complexity: \mathcal{O}(n+n^{2/3}\epsilon^{-1}) -> \mathcal{O}(n+n^{1/2}\epsilon^{-1}) | Helped |
| Use the SPIDER recursive gradient estimator instead of the SAGA-ADMM estimator | Nonconvex stochastic ADMM · Finite-sum nonconvex optimization | IFO complexity: \mathcal{O}(n+n^{2/3}\epsilon^{-1}) -> \mathcal{O}(n+n^{1/2}\epsilon^{-1}) | Helped |
| Use online SPIDER-ADMM instead of SADMM | Online nonconvex ADMM · Online nonconvex optimization | IFO complexity: \mathcal{O}(\epsilon^{-2}) -> \mathcal{O}(\epsilon^{-\frac{3}{2}}) | Helped |
| Use SPIDER-ADMM rather than nc-ADMM, nc-SVRG-ADMM, nc-SAGA-ADMM, or nc-SADMM | Nonconvex ADMM algorithms · Graph-guided binary classification and multi-task learning | Objective value versus CPU time | Helped |
Closest papers by meaning
- Faster and Non-ergodic O(1/K) Stochastic Alternating Direction Method of Multipliers (NeurIPS 2017 · similarity 0.91)
- A Communication Efficient Stochastic Multi-Block Alternating Direction Method of Multipliers (NeurIPS 2019 · similarity 0.91)
- ADMM for Nonconvex Optimization under Minimal Continuity Assumption (ICLR 2025 · similarity 0.90)
- ADMM without a Fixed Penalty Parameter: Faster Convergence with New Adaptive Penalization (NeurIPS 2017 · similarity 0.89)
- Accelerated Variance Reduced Stochastic ADMM (AAAI 2017 · similarity 0.89)
- SPIDER: Near-Optimal Non-Convex Optimization via Stochastic Path-Integrated Differential Estimator (NeurIPS 2018 · similarity 0.89)
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.