Research map / Federated learning and privacy / Communication-efficient FL
Client participation analysis
Prove convergence of federated averaging under client heterogeneity and arbitrary participation.
68 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2019–2026) · an approach in Communication-efficient FL, keyword cluster “optimization · convex · fedavg”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
31 papers in 2023–24 and 20 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 20.9% to 17.2%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2019 · Agnostic Federated Learning (ICML 2019 · 403 citations)
- 2020 · Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization (NeurIPS 2020 · 565 citations)
- 2021 · Toward Understanding the Influence of Individual Clients in Federated Learning (AAAI 2021 · 47 citations)
- 2022 · FedCor: Correlation-Based Active Client Selection Strategy for Heterogeneous Federated Learning (CVPR 2022 · 136 citations)
- 2023 · Faster Adaptive Federated Learning (AAAI 2023 · 86 citations)
- 2024 · Communication-Efficient Federated Learning with Accelerated Client Gradient (CVPR 2024 · 38 citations)
- 2025 · Widening the Network Mitigates the Impact of Data Heterogeneity on FedAvg (ICML 2025 · 0 citations)
- 2026 · Delayed Momentum Aggregation: Communication-efficient Byzantine-robust Federated Learning with Partial Participation (ICML 2026 · 0 citations)
Earliest papers here
- Agnostic Federated Learning (ICML 2019 · 403 citations)
- Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization (NeurIPS 2020 · 565 citations)
- Fair Resource Allocation in Federated Learning (ICLR 2020 · 351 citations)
What has been tried
50 papers here report 478 measured changes (a component added, removed or swapped, compared with a baseline): helped 68%, no clear effect 13%, hurt 15%, mixed 4%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add server update with momentum | CIFAR-10 and CIFAR-100, 500 clients, 2% participation, 1000R · accuracy | 71.45 -> 82.80; 38.11 -> 48.40 | Helped | Communication-Efficient Federated Learning with Accelerated Client Gradient (CVPR 2024 · 38 citations) |
| Omit the second SecureInnerProduct step | Hybrid FL experiments · empirical performance | No clear effect | A Primal-Dual Algorithm for Hybrid Federated Learning (AAAI 2024 · 10 citations) | |
| Decrease data similarity from s=100% to s=2.5% | Fashion-MNIST · test accuracy | 84.6% -> 84.45% | Hurt | Federated Learning under Periodic Client Participation and Heterogeneous Data: A New Communication-Efficient Algorithm and Analysis (NeurIPS 2024 · 8 citations) |
| Add the global scheduler to the client-side local scheduler | CIFAR10 and Shakespeare · performance | Helped | FedHyper: A Universal and Robust Learning Rate Scheduler for Federated Learning with Hypergradient Descent (ICLR 2024 · 1 citation) | |
| Use finite cutoff K=50 rather than K→∞ | SVHN test, Bernoulli · accuracy | 89.3{ ±0.5} -> 89.6{ ±0.4} | No clear effect | A Lightweight Method for Tackling Unknown Participation Statistics in Federated Averaging (ICLR 2024 · 0 citations) |
| increase server auxiliary data size from 50 to 1000 | MNIST, s=4, p=1 · test accuracy improvement (%) | 16.65 -> 31.07 | Helped | Understanding Server-Assisted Federated Learning in the Presence of Incomplete Client Participation (ICML 2024 · 0 citations) |
| Add effective learning-rate clipping | CIFAR-100 ResNet, keep ratio 0.1 · accuracy | 30.14 ± 1.58 -> 35.12 ± 1.51 | Helped | On Sampling Strategies for Spectral Model Sharding (NeurIPS 2024 · 0 citations) |
| add momentum-based stochastic control to FedSWA | CIFAR100 ResNet-18, Dirichlet-0.6, 10% client participation · accuracy | 59.8 -> 67.9 | Helped | Improving Generalization in Federated Learning with Highly Heterogeneous Data via Momentum-Based Stochastic Controlled Weight Averaging (ICML 2025) |
What reviewers will expect
Datasets and baselines named in the 12 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- CIFAR-10 · 7 of 12 papers
- MNIST · 3 of 12 papers
- Tiny ImageNet · 2 of 12 papers
- FMNIST · 2 of 12 papers
Baselines
- SCAFFOLD · 4 of 12 papers
- FedAvg · 4 of 12 papers
- FedNova · 2 of 12 papers
- MIFA · 2 of 12 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 51 papers; from OpenAlex and the papers' affiliations).
- The Chinese University of Hong Kong · 5 papers
- University of Maryland, College Park · 4 papers
- Google · 3 papers
- The Hong Kong Polytechnic University · 2 papers
- University of Pittsburgh · 2 papers
- University of Michigan · 2 papers
- The Ohio State University · 2 papers
- Technical University of Denmark · 2 papers
- CISPA Helmholtz Center for Information Security · 2 papers
- Meta · 2 papers
Use it
Download BibTeX of 38 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Federated optimization algorithms (70 papers)
- Local drift correction (61 papers)
Elsewhere in the topic
- Efficient local training in FL (77 papers)
- Communication-efficient heterogeneous FL (65 papers)
- Decentralized peer-to-peer training (41 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).