RCFL: Redundancy-Aware Collaborative Federated Learning in Vehicular Networks
Bibliographic record
Abstract
In vehicular networks (VNets), vehicular federated learning (VFL) is a new learning paradigm that can protect data privacy of vehicle nodes (VNs) while training models. In VFL, the importance of data (IoD) is a key factor that affects model training accuracy. However, due to the heterogeneity of data in the VFL, it is a challenge to evaluate the quality of data owned by different VNs and design an efficient federated learning scheme to enable the VNs to complete learning tasks collaboratively. In this paper, we consider the IoD and propose a redundancy-aware collaborative federated learning (RCFL) scheme for the VFL. In the scheme, by jointly considering the data quality and the cooperation among VNs, we first design a redundancy-aware federated learning architecture to efficiently provide learning services in VNets. Then, we develop a data importance model that integrates the non-independent and identically distributed (non-IID) degree and the redundancy of data (RoD) to evaluate the data quality and formulate the cooperation of the VNs as a coalition game to improve their data importance, where the equilibrium of the coalition game is obtained by designing a coalition formation algorithm. After that, by considering the diversified characteristics of data and the available resources of different VNs in each coalition, a coalition-based federated learning algorithm is designed to enable the distributed coalitions to complete the learning task cooperatively with the target of improving the learning accuracy. The simulation results show that the proposed scheme outperforms the benchmark schemes in terms of the IoD obtained by the VNs and the training accuracy.
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 | 0.006 |
| 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 | 0.003 | 0.003 |
| 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".