Research map / Optimization / Nonconvex and smooth optimization
Nonconvex Zeroth-Order Stochastic ADMM Methods with Lower Function Query Complexity
Feihu Huang, Shangqian Gao, Jian Pei, Heng Huang
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 ZO-SPIDER-ADMM instead of ZO-SAGA-ADMM, ZO-SVRG-ADMM and ZO-SGD-ADMM | zeroth-order stochastic ADMM · universal structured black-box attack, finite-sum | attack loss vs epochs | Improved |
| use ZOO-ADMM+ instead of ZOO-ADMM and ZO-GADM | zeroth-order online ADMM · universal structured black-box attack, online | attack loss vs iterations | Improved |
| use the mixed CooGE+UniGE estimator instead of CooGE alone in ZO-SPIDER-ADMM (ablation) | ZO-SPIDER-ADMM · black-box attack, finite-sum | attack loss vs number of queries | Improved |
| use the mixed CooGE+UniGE estimator instead of CooGE alone in ZOO-ADMM+ (ablation) | ZOO-ADMM+ · black-box attack, online | attack loss vs number of queries | Improved |
| use expensive CooGE estimator (2d queries per gradient) alone (ablation) | ZOO-ADMM+ · black-box attack, online | attack loss vs number of queries | Hurt |
| change target model from 7-layer DNN to ResNet-20 | ZOO/ZO-SPIDER ADMM · CIFAR-10 black-box attack | attack loss vs number of queries | No clear effect |
Closest papers by meaning
- Faster Stochastic Alternating Direction Method of Multipliers for Nonconvex Optimization (ICML 2019 · similarity 0.91)
- Improved Zeroth-Order Variance Reduced Algorithms and Analysis for Nonconvex Optimization (ICML 2019 · similarity 0.90)
- Accelerated Stochastic Gradient-free and Projection-free Methods (ICML 2020 · similarity 0.89)
- ADMM for Nonconvex Optimization under Minimal Continuity Assumption (ICLR 2025 · similarity 0.87)
- Stochastic Zeroth-Order Optimization Under Heavy-Tailed Noise (NeurIPS 2026 · similarity 0.86)
- Zeroth-Order Optimization Finds Flat Minima (NeurIPS 2025 · 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.
