AI Paper Atlas

Research map / Federated learning and privacy / Communication-efficient FL

Federated Multi-Objective Learning

Haibo Yang, Zhuqing Liu, Jia Liu, Chaosheng Dong, Michinari Momma

NeurIPS 2023 · Federated learning and privacy › Communication-efficient FL · approach: Federated optimization algorithms · 3 citations (OpenAlex)

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
increase local update steps K from 1 to 20 (ablation)FMGDA/FSMGDA · MultiMNIST i.i.d. Task Lcommunication rounds to 1e-2 loss: 82 -> 5Improved
increase local update steps K from 1 to 20 (ablation)FMGDA/FSMGDA · MultiMNIST non-i.i.d. Task Lcommunication rounds to 1e-2 loss: 96 -> 6Improved
increase local update steps K from 1 to 20 (ablation)FMGDA/FSMGDA · MultiMNIST non-i.i.d. Task Rcommunication rounds to 1e-2 loss: 82 -> 5Improved
increase local update steps K from 1 to 20 (ablation)FMGDA/FSMGDA · MultiMNIST i.i.d. Task Rcommunication rounds to 1e-2 loss: 84 -> 5Improved
increase local steps K from 1 to 10 (ablation)FMGDA/FSMGDA · MultiMNIST i.i.d. Task Rcommunication rounds to 1e-2 loss: 84 -> 9Improved
increase local steps K from 1 to 5 (ablation)FMGDA/FSMGDA · MultiMNIST non-i.i.d. Task Lcommunication rounds to 1e-2 loss: 96 -> 24Improved
increase batch size from 16 to 256 (256 = full-batch FMGDA) (ablation)FSMGDA · MultiMNIST non-i.i.d. training lossconvergence speedImproved
federated FMGDA versus centralized MGDFMGDA · MultiMNIST training loss after 100 roundstraining lossNo clear effect
federated FSMGDA versus centralized SMGDFSMGDA · MultiMNIST training loss after 100 roundstraining lossHurt
more local steps K and larger batch size (ablation)FMGDA/FSMGDA · River Flow (8 tasks)normalized loss convergenceImproved

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