Energy-Adaptive Multi-Dimensional Learning Control for Federated Learning in Energy-Harvesting AIoT Systems
Dong Kun Noh, Changmin Kwak · 2026
Read Energy-Adaptive Multi-Dimensional Learning Control for Federated Learning in Energy-Harvesting AIoT Systems by Dong Kun Noh, Changmin Kwak (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13259020.txt.
Description
Dong Kun Noh and Changmin Kwak control shared model training on solar powered devices in a 2026 study. Devices that harvest energy can black out mid training when batteries drain. Their controller adjusts model size and training effort to match each device energy level in real time.
Devices join training rounds only when stored energy clears a threshold. Joining devices pick settings covering pruning, compression, and local workload that fit their energy budget. Trials on small edge computers under solar harvesting cut blackouts sharply. Accuracy stayed competitive with setups that ignored energy limits. Designers of off grid sensor networks gain a practical training controller.
Catalog record
| CREATOR | Dong Kun Noh, Changmin Kwak |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |