FL-EGM: Decentralized Federated Learning using Aggregator Selection with Enhanced Global Model
Bibliographic record
Abstract
Federated Learning (FL) has emerged as a promising solution to address data privacy concerns, but it also faces ob-stacles, including biased centralized aggregation. In this paper, we propose a novel decentralized FL architecture with round-based aggregator selection that enhances the accuracy of the global model while preserving data privacy for the users. Our proposed model consists of three distinct phases. Firstly, we address the issue of a central aggregator by introducing a model where we periodically select the aggregator as the best-performing client from participant nodes. Secondly, we train all clients using their respective data but excluding the client selected as an aggregator. In the third phase, we further refine the global model by training it on the raw data of the selected aggregator, leading to an enhanced global model. These three modules empower the proposed model to offer improved accuracy and decentralization. Our implementation yielded promising results, demonstrating that the proposed mode achieves an accuracy of 98.54 % along with enhanced decentralization. The proposed model has also validated different datasets and network conditions, such as the number of participant nodes. Furthermore, the performance of the proposed model is validated in the presence of a biased aggregator. The results demonstrate that the proposed model achieves 98.43 % accuracy and shows more robustness in the presence of a biased aggregator. Finally, the proposed model converges fast compared to other baseline models.
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 | 0.002 | 0.003 |
| 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 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".