Decentralized Gossip Learning with Adaptive OOD Detection for LEO Satellite Networks under Non-IID Data
Bibliographic record
Abstract
Low Earth Orbit (LEO) satellite constellations offer great promise for distributed sensing and intelligence in future networks. However, their potential is hindered by communication inefficiencies and heterogeneous data distributions, which lead to excessive transmission costs, slow convergence, and degraded model performance. Existing federated learning (FL) methods alleviate data privacy and communication load, but typically rely on centralized coordination and lack robustness to Non-IID data, especially in dynamic satellite environments. To address these limitations, we propose the GOOD (Gossip Learning with OOD Enhancement) framework, a fully decentralized, gossip-based FL system enhanced with adaptive out-of-distribution (OOD) detection. GOOD enables satellites to evaluate the distributional compatibility of incoming model updates using lightweight OOD scores, thereby reducing negative transfer and eliminating ineffective transmissions. Additionally, it introduces an adaptive grouping mechanism that dynamically clusters satellites based on distributional similarity, optimizing inter-satellite communication. Extensive experiments demonstrate that GOOD consistently outperforms prior decentralized FL approaches in terms of convergence accuracy and communication efficiency under both statistical and semantic Non-IID conditions, making it a compelling solution for autonomous on-orbit learning.
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 | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| 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".