Research map / Federated learning and privacy / Communication-efficient FL
Federated Submodel Optimization for Hot and Cold Data Features
Yucheng Ding, Chaoyue Niu, Fan Wu, Shaojie Tang, Chengfei Lyu, yanghe feng, Guihai Chen
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 FedSubAvg instead of FedAvg | FedAvg · MovieLens LR | rounds to target loss: 170 -> 100 | Improved |
| use FedSubAvg instead of CentralSGD | CentralSGD · Sent140 LSTM | rounds to target loss: 980 -> 260 | Improved |
| use FedSubAvg instead of FedAvg | FedAvg · Sent140 LSTM | rounds to target loss: 1,000+ -> 260 | Improved |
| use FedSubAvg instead of FedAvg | FedAvg · Amazon DIN | highest test AUC: 0.523 -> 0.641 | Improved |
| use FedSubAvg instead of FedAdam | FedAdam · Amazon DIN | rounds to target AUC: 16 -> 53 | Hurt |
| use FedSubAvg instead of FedAdam | FedAdam · Amazon DIN | highest test AUC: 0.633 -> 0.641 | Improved |
| use FedSubAvg instead of FedAvg | FedAvg · Alibaba DIN | highest test AUC: 0.509 -> 0.626 | Improved |
| use FedSubAvg instead of FedAdam | FedAdam · Alibaba DIN | highest test AUC: 0.554 -> 0.626 | Improved |
| compare FedSubAvg with centralized SGD | CentralSGD · Alibaba DIN | rounds to target AUC: 265 -> 610 | Mixed |
| increase selected clients per round K from 10 to 50 (ablation) | FedSubAvg · Sent140 LSTM | rounds to target loss: 420 -> 260 | Improved |
| increase selected clients per round K from 20 to 100 (ablation) | FedSubAvg · Amazon DIN | rounds to target AUC: 95 -> 53 | Improved |
| increase selected clients per round K from 20 to 100 (ablation) | FedSubAvg · Alibaba DIN | rounds to target AUC: 3,469 -> 610 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Scaffold — SCAFFOLD: Stochastic Controlled Averaging for Federated Learning (ICML 2020) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 2025 · similarity 0.88)
- FedSMU: Communication-Efficient and Generalization-Enhanced Federated Learning through Symbolic Model Updates (ICML 2025 · similarity 0.88)
- Improving Generalization in Federated Learning with Highly Heterogeneous Data via Momentum-Based Stochastic Controlled Weight Averaging (ICML 2025 · similarity 0.87)
- FedImpro: Measuring and Improving Client Update in Federated Learning (ICLR 2024 · similarity 0.87)
- FedLLM-Bench: Realistic Benchmarks for Federated Learning of Large Language Models (NeurIPS 2024 · similarity 0.87)
- Decomposable Submodular Maximization in Federated Setting (ICML 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.
