Research map / Optimization / Distributed and compressed
A Decentralized Parallel Algorithm for Training Generative Adversarial Nets
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 |
|---|---|---|---|
| Replace centralized all-reduce with decentralized ring communication | DP-OAdam · CIFAR10 | Inception Score convergence by epoch | No clear effect |
| Replace centralized all-reduce with decentralized ring communication | DP-OAdam · ImageNet | Inception Score convergence by epoch | No clear effect |
| Use random neighbor mixing instead of a fixed ring topology (ablation) | DP-OAdam · CIFAR10 and ImageNet | Inception Score convergence by epoch | Helped |
| Use decentralized communication instead of centralized all-reduce | DP-OAdam and Rand-DP-OAdam · CIFAR10 and ImageNet, HPC environment | speedup | Helped |
| Use decentralized communication instead of centralized all-reduce | DP-OAdam and Rand-DP-OAdam · CIFAR10 and ImageNet, high-latency cloud environment | run-time | Helped |
| Use random neighbor mixing instead of a fixed ring topology (ablation) | DP-OAdam · CIFAR10 and ImageNet, HPC environment | speedup | Helped |
Closest papers by meaning
- From Promise to Practice: Realizing High-performance Decentralized Training (ICLR 2025 · similarity 0.91)
- Consensus Control for Decentralized Deep Learning (ICML 2021 · similarity 0.88)
- $\textbf{A}^2\textbf{CiD}^2$: Accelerating Asynchronous Communication in Decentralized Deep Learning (NeurIPS 2023 · similarity 0.88)
- TRAINING GENERATIVE ADVERSARIAL NETWORKS VIA PRIMAL-DUAL SUBGRADIENT METHODS: A LAGRANGIAN PERSPECTIVE ON GAN (ICLR 2018 · similarity 0.87)
- Finding Mixed Nash Equilibria of Generative Adversarial Networks (ICML 2019 · similarity 0.87)
- DIMAT: Decentralized Iterative Merging-And-Training for Deep Learning Models (CVPR 2024 · similarity 0.87)
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.