Research map / Optimization / Nonconvex and smooth optimization
Asynchronous Mini-Batch Gradient Descent with Variance Reduction for Non-Convex 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 |
|---|---|---|---|
| Add variance reduction to asynchronous SGD | AsySVRG · Non-convex smooth optimization | convergence rate: O(1/\sqrt{T}) -> O(1/T) | Helped |
| Use SVRG rather than SGD | MNIST · MNIST training | training loss convergence | Mixed |
| Initialize SVRG from an SGD-pretrained model | MNIST · MNIST | convergence rate | Helped |
| Use SVRG rather than SGD | CIFAR-10 · CIFAR-10 training | training loss convergence | Mixed |
| Initialize SVRG from an SGD-pretrained model | CIFAR-10 · CIFAR-10 | training convergence rate | Helped |
| Increase the number of workers | Shared-AsySVRG · MNIST | running time speedup | Helped |
| Increase the number of workers | Distributed-AsySVRG · CIFAR-10 | running time speedup | Helped |
| Increase worker count beyond a threshold | Shared-AsySVRG · MNIST | running time speedup | Hurt |
| Increase worker count substantially | Distributed-AsySVRG · CIFAR-10 | running time speedup | Hurt |
| Use SGD-pretrained initialization instead of the untrained initialization | CIFAR-10 SVRG · CIFAR-10 test set | test error | No clear effect |
Closest papers by meaning
- Parallel Asynchronous Stochastic Variance Reduction for Nonconvex Optimization (AAAI 2017 · similarity 0.93)
- Fast Asynchronous Parallel Stochastic Gradient Descent: A Lock-Free Approach with Convergence Guarantee (AAAI 2016 · similarity 0.93)
- Sharper Convergence Guarantees for Asynchronous SGD for Distributed and Federated Learning (NeurIPS 2022 · similarity 0.91)
- Ringmaster ASGD: The First Asynchronous SGD with Optimal Time Complexity (ICML 2025 · similarity 0.89)
- Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity (NeurIPS 2026 · similarity 0.89)
- Stability and Generalization of Asynchronous SGD: Sharper Bounds Beyond Lipschitz and Smoothness (NeurIPS 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.