What changes after deployment? A survey on On-device Learning in TinyML
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911732125925376 |
|---|---|
| author | Pavan, Massimo Pezzarossa, Luca Pittorino, Fabrizio Roveri, Manuel Fafoutis, Xenofon |
| author_facet | Pavan, Massimo Pezzarossa, Luca Pittorino, Fabrizio Roveri, Manuel Fafoutis, Xenofon |
| contents | Machine learning models on microcontroller-class devices (TinyML) face a fundamental challenge: post-deployment distribution change undermines static models. On-device learning (ODL) addresses this by running the learning process directly on the device. The existing literature has not characterized how distribution change occurs or how different change types require different solutions. Approximately 70 ODL works are surveyed under one principle: the distribution change regime. The survey analyzes how different types of distribution change influence the applications addressable on-device, the hardware employed, and the structure of the solutions. A persistent gap between methodological benchmarks and real-world deployment scenarios is also identified. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_31226 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | What changes after deployment? A survey on On-device Learning in TinyML Pavan, Massimo Pezzarossa, Luca Pittorino, Fabrizio Roveri, Manuel Fafoutis, Xenofon Machine Learning Artificial Intelligence I.2.6 Machine learning models on microcontroller-class devices (TinyML) face a fundamental challenge: post-deployment distribution change undermines static models. On-device learning (ODL) addresses this by running the learning process directly on the device. The existing literature has not characterized how distribution change occurs or how different change types require different solutions. Approximately 70 ODL works are surveyed under one principle: the distribution change regime. The survey analyzes how different types of distribution change influence the applications addressable on-device, the hardware employed, and the structure of the solutions. A persistent gap between methodological benchmarks and real-world deployment scenarios is also identified. |
| title | What changes after deployment? A survey on On-device Learning in TinyML |
| topic | Machine Learning Artificial Intelligence I.2.6 |
| url | https://arxiv.org/abs/2605.31226 |