Research map / Federated learning and privacy / Communication-efficient FL
Towards Instance-adaptive Inference for Federated Learning
Chun-Mei Feng, Kai Yu, Nian Liu, Xinxing Xu, Salman Khan, Wangmeng Zuo
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 |
|---|---|---|---|
| add instance-adaptive SSF pool to federated SSF (ablation) | Federated SSF · Tiny-ImageNet | accuracy: 84.32 -> 86.79 | Improved |
| use SSF instead of prompt tuning (ablation) | FedIns · Tiny-ImageNet | accuracy: 81.02 -> 84.32 | Improved |
| add a pool to prompt tuning (ablation) | Prompt · Tiny-ImageNet | accuracy: 81.02 -> 81.98 | Improved |
| FedIns instead of FedAvg full fine-tuning at 15% communication cost | FedAvg · Tiny-ImageNet | accuracy: 79.02 -> 86.79 | Improved |
| use FedProx and SCAFFOLD instead of FedAvg with little local data | FedAvg · CIFAR-100 | accuracy: 81.22 -> 78.02 (SCAFFOLD) | Hurt |
| FedDC instead of FedAvg | FedAvg · CIFAR-100 | accuracy: 81.22 -> 78.44 | Hurt |
| increase pool size M up to 25 (ablation) | FedIns · DomainNet, CIFAR-100, Tiny-ImageNet | accuracy | Improved |
| increase pool size beyond 25 (ablation) | FedIns · three datasets | accuracy | Hurt |
| increase best-matched number C beyond 3 (ablation) | FedIns · CIFAR-100 | accuracy | Hurt |
| increase C on Tiny-ImageNet (ablation) | FedIns · Tiny-ImageNet | accuracy | No clear effect |
| increase intra-client heterogeneity to gamma=20 | FedIns vs FedBN · Tiny-ImageNet | accuracy: FedIns 75.63 -> 64.15; FedBN 68.53 -> 31.69 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- FedDC — FedDC: Federated Learning With Non-IID Data via Local Drift Decoupling and Correction (CVPR 2022) · better on 1 of 1 shared result
- SCAFFOLD — SCAFFOLD: Stochastic Controlled Averaging for Federated Learning (ICML 2020) · better on 1 of 1 shared result
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
- Federated Learning with Sample-level Client Drift Mitigation (AAAI 2025 · similarity 0.90)
- FedGPS: Statistical Rectification Against Data Heterogeneity in Federated Learning (NeurIPS 2025 · similarity 0.90)
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 2025 · similarity 0.90)
- FedSMU: Communication-Efficient and Generalization-Enhanced Federated Learning through Symbolic Model Updates (ICML 2025 · similarity 0.90)
- Improving Generalization in Federated Learning with Model-Data Mutual Information Regularization: A Posterior Inference Approach (NeurIPS 2024 · similarity 0.89)
- An Aggregation-Free Federated Learning for Tackling Data Heterogeneity (CVPR 2024 · similarity 0.89)
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.
