Non-Cooperative Edge Server Selection Game for Federated Learning in IoT
Bibliographic record
Abstract
Computational offloading is an efficient way to help constrained IoT devices by performing heavy tasks on Edge servers, especially tasks related to Machine Learning. Moreover, due to their limited learning capacity and memory size, such devices can only store a limited amount of data as a training set for their learning. Consequently, learning prediction is bound to be smeared with relatively high error. To mend that issue, IoT devices can federate the learning process with their pairs via an Edge server. However, offloading repeatedly the learning model through a wireless access network is time consuming. Hence, although learning collectively can reduce the learned model variance, it inflicts a communication cost depending on the selected Edge server. Therefore, in this paper, we model the Edge Selection problem as a non-cooperative game where devices autonomously and efficiently select an Edge server to reduce both their learning error and their communication cost. Depending on the characteristics of the dataset, we discern two different types of games. For each game type, we implemented and compared a semi-distributed algorithm based on Best Response dynamics. We compared the obtained results with the optimal centralized approach and with a less computationally intensive meta-heuristics, to assess the price of anarchy. Our numerical analysis shows that the Best Response algorithm strikes a good balance between efficiency and swift convergence.
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 | 0.000 | 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) | 0.002 | 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".