Metraipolis › archive › dc_fa3d437c009f

16-bit Precision of Convolutional Neural Networks on Microcontroller Units for 8-bit Costs — arXiv:2610.03402

Source
arXiv
Original
https://arxiv.org/abs/2610.03402
License (as archived)
CC0-1.0
Archived
2026-10-05T23:10:16.714Z
SHA-256 of the archived text
74d9f37de78b0e6a2f25d0b07a3070e455915829410db4c58b7aadbd1b9418b7
Stable URL of this page
https://metraipolis.com/documento/dc_fa3d437c009f

How to cite

16-bit Precision of Convolutional Neural Networks on Microcontroller Units for 8-bit Costs — arXiv:2610.03402 — arXiv, https://arxiv.org/abs/2610.03402 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fa3d437c009f (Metraipolis archive dc_fa3d437c009f, sha256 74d9f37de78b0e6a2f25d0b07a3070e455915829410db4c58b7aadbd1b9418b7)
BibTeX
@misc{dcfa3d437c009f,
  title = {16-bit Precision of Convolutional Neural Networks on Microcontroller Units for 8-bit Costs — arXiv:2610.03402},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.03402},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fa3d437c009f; sha256 74d9f37de78b0e6a2f25d0b07a3070e455915829410db4c58b7aadbd1b9418b7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03402. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

To deploy deep neural networks on edge hardware, highly efficient inference schemes are necessary that retain high accuracy. This work presents W16A16, a high precision (16-bit), fast speed, low energy quantization method. On a widely applied microcontroller architecture Armv7E-M, our proposed approach achieves faster speed and lower energy consumption on layer- and model-level compared to alternative quantization schemes. We analyze the architecture of Armv7E-M, explain the underlying principles behind the performance advantages of 16-bit approaches, and evaluate the empiric quantization errors for regression and classification tasks, as well as empiric time- and energy consumption in MCU deployment. We observe ca.\ 10 times lower quantization errors compared to 8-bit quantization schemes while achieving similar or better inference times and energy consumption.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.03402 — 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 2026-10-05T23:01:38.692Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).