A Dual-Component Personalization Strategy for UShaped Split Federated Learning
Bibliographic record
Abstract
This work investigates the personalization capabilities of U-shaped Split Federated Learning (U-SFL) in statistically heterogeneous environments, challenging the default configurations where client-side components are treated as either fully global or fully local. To explore the performance landscape between these poles, a dual-component personalization strategy using model interpolation is introduced, controlled by two independent interpolation factors for the client-side head and tail components . Through experiments on the CIFAR-10 dataset under a non-IID data distribution, the influence of these parameters on both personalization and generalization was evaluated by varying each factor independently while holding the other constant at a baseline. The empirical findings demonstrate that a refined partial personalization method outperforms the baseline models in both personalization and generalization accuracy. The optimal configuration was found to be a fully global head combined with a partially personalized tail. The findings demonstrate that the degree of personalization in U-SFL can be treated as a tunable hyperparameter, shedding light on an effective and balanced training approach for non-IID settings.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".