Research map / Optimization / Distributed and compressed
3PC: Three Point Compressors for Communication-Efficient Distributed Training and a Better Theory for Lazy Aggregation
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 |
|---|---|---|---|
| Combine lazy aggregation with Top-K compression | CLAG · ijcnn1 logistic regression | communication complexity to reach gradient norm below 10^-2 | Helped |
| Combine lazy aggregation with Top-K compression | CLAG · LIBSVM phishing, w6a, a9a, and ijcnn1 | communication complexity | Helped |
| Use 3PCv2 with alternative first compressors, including Perm-K and Rand-K, while keeping Top-K second | 3PCv2 · MNIST autoencoder, especially heterogeneous data | squared gradient norm | Mixed |
| Use an adaptive three-point compressor whose compression error decays over training | 3PC · Smooth nonconvex distributed optimization | convergence rate | Helped |
| Use the three-point compressor framework for lazy aggregation | LAG and CLAG · Smooth nonconvex distributed optimization | convergence rate | Helped |
Closest papers by meaning
- Improving the Worst-Case Bidirectional Communication Complexity for Nonconvex Distributed Optimization under Function Similarity (NeurIPS 2024 · similarity 0.88)
- Accelerated Methods with Compressed Communications for Distributed Optimization Problems Under Data Similarity (AAAI 2025 · similarity 0.86)
- LoCoDL: Communication-Efficient Distributed Learning with Local Training and Compression (ICLR 2025 · similarity 0.86)
- Beyond Communication Overhead: A Multilevel Monte Carlo Approach for Mitigating Compression Bias in Distributed Learning (ICML 2025 · similarity 0.86)
- Approximate Gradient Coding for Distributed Learning with Heterogeneous Stragglers (NeurIPS 2025 · similarity 0.85)
- Towards Faster Decentralized Stochastic Optimization with Communication Compression (ICLR 2025 · similarity 0.85)
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.