Metraipolis › archive › dc_93ae3c9d752c

Design-Time Optimization of Deep Neural Networks for Intermittent Learning on Microcontrollers — arXiv:2608.03589

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.03589
License (as archived)
CC0-1.0
Archived
2026-10-31T12:42:27.794Z
SHA-256 of the archived text
77a8f6d68ef9ce5c535f9807dfd7b7ce38dba94bc7c1b4beb8f31fe7f34c490a
Stable URL of this page
https://metraipolis.com/documento/dc_93ae3c9d752c

How to cite

Design-Time Optimization of Deep Neural Networks for Intermittent Learning on Microcontrollers — arXiv:2608.03589 — arXiv (en), https://arxiv.org/abs/2608.03589 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_93ae3c9d752c (Metraipolis archive dc_93ae3c9d752c, sha256 77a8f6d68ef9ce5c535f9807dfd7b7ce38dba94bc7c1b4beb8f31fe7f34c490a)
BibTeX
@misc{dc93ae3c9d752c,
  title = {Design-Time Optimization of Deep Neural Networks for Intermittent Learning on Microcontrollers — arXiv:2608.03589},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.03589},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_93ae3c9d752c; sha256 77a8f6d68ef9ce5c535f9807dfd7b7ce38dba94bc7c1b4beb8f31fe7f34c490a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03589. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present a method for designing deep neural networks (DNNs) for intermittent, energy-autonomous, on-device learning on microcontroller units (MCUs). In mobile applications where the energy can run out, e.g., when solar-powered, executing artificial intelligence (AI) faces a technical issue as learning can be interrupted at any time. Our approach combines a hardware-aware energy prediction model with multi-objective optimization (MOO), enabling offline DNN optimization at the design stage without repeated deployment and online testing on the target MCU. Our proposed energy predictor estimates per-layer energy consumption for both DNN inference and training, including the intermittent checkpointing overhead, based on implementation-specific compute and memory features extracted from the DNN model. We validate our approach using autoencoders for anomaly detection on a Cortex-M4 MCU, where our predictor achieves a weighted absolute percentage error of 16.6%, which is sufficient for reliable architecture selection under intermittency constraints. As a result, this work bridges the gap between MOO, automated DNN design, deployment on energy-harvesting systems, and intermittent learning, truly enabling autonomous AI at the edge.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.03589 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-31T12:31:30.360Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).