Research map / Optimization / Nonconvex and smooth optimization
D-SPIDER-SFO: A Decentralized Optimization Algorithm with Faster Convergence Rate for Nonconvex Problems
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 D-SPIDER-SFO instead of D-PSGD | Decentralized nonconvex optimization · Nonconvex decentralized optimization | gradient computation cost: \mathcal{O}(\epsilon^{-4}) -> \mathcal{O}( \epsilon^{-3}) | Helped |
| Use D-SPIDER-SFO instead of D$^2$ | Decentralized nonconvex optimization · Nonconvex decentralized optimization | gradient computation cost: \mathcal{O}(\epsilon^{-4}) -> \mathcal{O}( \epsilon^{-3}) | Helped |
| Use decentralized rather than centralized SPIDER-SFO communication | SPIDER-SFO · Nonconvex decentralized optimization | communication cost on the busiest node: \mathcal{O}(n) -> \mathcal{O} ( (Deg(network)) | Helped |
| Use D-SPIDER-SFO instead of C-SPIDER-SFO as communication becomes a bottleneck | LeNet5 on CIFAR-10, 8 GPUs · CIFAR-10 | wall clock time per epoch: one order of magnitude faster | Helped |
| Use D-SPIDER-SFO instead of D$^2$ with worker-specific class data | LeNet5 on CIFAR-10; workers each access two classes · CIFAR-10 unshuffled case | computational complexity | Helped |
| Use D-SPIDER-SFO rather than D-PSGD with worker-specific class data | LeNet5 on CIFAR-10; workers each access two classes · CIFAR-10 unshuffled case | convergence | Helped |
Closest papers by meaning
- An Online Optimization Perspective on First-Order and Zero-Order Decentralized Nonsmooth Nonconvex Stochastic Optimization (ICML 2024 · similarity 0.91)
- Decentralized Sum-of-Nonconvex Optimization (AAAI 2024 · similarity 0.90)
- Faster Adaptive Decentralized Learning Algorithms (ICML 2024 · similarity 0.89)
- Decentralized Convex Finite-Sum Optimization with Better Dependence on Condition Numbers (ICML 2024 · similarity 0.89)
- Jointly Improving the Sample and Communication Complexities in Decentralized Stochastic Minimax Optimization (AAAI 2024 · similarity 0.88)
- Decentralized Gradient-Free Methods for Stochastic Non-smooth Non-convex Optimization (AAAI 2024 · similarity 0.88)
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.