Pith. sign in

REVIEW 5 cited by

FedBABU: Towards Enhanced Representation for Federated Image Classification

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2106.06042 v3 pith:MPALZOOE submitted 2021-06-04 cs.LG

Jaehoon Oh , Sangmook Kim , Se-Young Yun This is my paper · ORCID
classification cs.LG
keywords federatedpersonalizationfedbabuheadmodelperformancebodydata
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Federated learning has evolved to improve a single global model under data heterogeneity (as a curse) or to develop multiple personalized models using data heterogeneity (as a blessing). However, little research has considered both directions simultaneously. In this paper, we first investigate the relationship between them by analyzing Federated Averaging at the client level and determine that a better federated global model performance does not constantly improve personalization. To elucidate the cause of this personalization performance degradation problem, we decompose the entire network into the body (extractor), which is related to universality, and the head (classifier), which is related to personalization. We then point out that this problem stems from training the head. Based on this observation, we propose a novel federated learning algorithm, coined FedBABU, which only updates the body of the model during federated training (i.e., the head is randomly initialized and never updated), and the head is fine-tuned for personalization during the evaluation process. Extensive experiments show consistent performance improvements and an efficient personalization of FedBABU. The code is available at https://github.com/jhoon-oh/FedBABU.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 48 citations worldwide. Full citation record

  1. PLayer-FL: A Principled Approach to Personalized Layer-wise Cross-Silo Federated Learning

    cs.LG 2025-02 conditional novelty 5.0 of 10

    PLayer-FL picks the layer split in partial federated learning from a low-cost sensitivity metric computed at epoch 1, and reports competitive F1, fairness, and participation incentives across seven non-IID datasets.

  2. UniVarFL: Uniformity and Variance Regularized Federated Learning for Heterogeneous Data

    cs.LG 2025-06 reject novelty 4.0 of 10

    UniVarFL adds a classifier variance regularizer and a hyperspherical uniformity regularizer to local federated training, reporting improved accuracy on some non-IID benchmarks but not consistently across its own experiments.

  3. pFedSOP : Accelerating Training Of Personalized Federated Learning Using Second-Order Optimization

    cs.DC 2025-06 reject novelty 4.0 of 10

    pFedSOP combines Gompertz-weighted local/global gradients with a rank-one Fisher Information Matrix update to speed up personalized federated learning, but the convergence proof is invalid and the update reduces to no...

  4. PFedDST: Personalized Federated Learning with Decentralized Selection Training

    cs.LG 2025-02 reject novelty 4.0 of 10

    A decentralized personalized federated learning method that scores peers by loss, header similarity, and recency reports faster convergence, but its own CIFAR-100 result contradicts the accuracy claim.

  5. Personalized Federated Learning via Dual-Prompt Optimization and Cross Fusion

    cs.LG 2025-06 conditional novelty 3.0 of 10

    pFedDC combines global and local text and vision prompts with cross-attention fusion to personalize federated CLIP models under label and domain shift.

Pith tools