AI Paper Atlas

Research map / Federated learning and privacy / Personalized and heterogeneous FL

AnaPFL: When Closed-Form Solutions Meet Generalizationand Personalization in Personalized Federated Learning

Kejia Fan, Jianheng Tang, Zhirui Yang, Feijiang Han, Yajiang Huang, Run He, Jiaxu Li, Songning Lai, Anfeng Liu, Houbing Herbert Song, Yunhuai Liu, HUIPING ZHUANG

ECCV 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Personalization with global-local trade-off

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 changedWhereMeasuredEffect
add dual-stream (primary + refinement) analytic personalization to AFLAFL · CIFAR-100, 50 clients, alpha=0.1average accuracy: 56.63% -> 80.64%Improved
add refinement stream for local personalization to AFLAFL · CIFAR-100, 50 clients, alpha=1.0average accuracy: 56.66% -> 62.63%Improved
add refinement stream to AFL on ImageNet-RAFL · ImageNet-R, 50 clients, alpha=0.1average accuracy: 28.87% -> 46.23%Improved
analytic AnaPFL instead of gradient-based FedSelectFedSelect · ImageNet-R, 50 clients, alpha=0.1average accuracy: 34.83% -> 46.23%Improved
increase clients from 50 to 100AnaPFL · CIFAR-100, alpha=0.1average accuracy: 80.64% -> 79.22%Hurt
lower heterogeneity alpha from 0.1 to 1.0AnaPFL · CIFAR-100, 50 clientsaverage accuracy: 80.64% -> 62.63%Hurt
add local fine-tuning (personalization) to FedAvg under high heterogeneityFedAvg · CIFAR-100, 50 clients, alpha=0.1average accuracy: 43.53% -> 58.81%Improved
add local fine-tuning to FedAvg under low heterogeneityFedAvg · CIFAR-100, 50 clients, alpha=1.0average accuracy: 48.91% -> 37.73%Hurt
use analytic learning for aggregation instead of gradient-based updates in efficiency comparisonFedAvg and FedPCL · CIFAR-100, 50 clients, alpha=0.1computation time: 100-300 s vs 45,000-66,000 sImproved
use analytic learning instead of gradient-based updates for communicationFedAvg and FedPCL · CIFAR-100, 50 clients, alpha=0.1communication overhead: 200-400 MB vs 20,000-40,000 MBImproved
add personalization to AFL costs slight extra overheadAFL · CIFAR-100efficiency: accuracy gaps of up to 17.36%-24.01%Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most 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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy