Skip to content
CaMLSysUniversity of Cambridge
Home People Publications Projects Contact

← All publications

Neural networks on microcontrollers: saving memory at inference via operator reordering

  • Edgar LiberisUniversity of Oxford
  • Nicholas D. LaneUniversity of OxfordSamsung AI Center, Cambridge

Preprint arXiv, 2019

Abstract

Designing deep learning models for highly-constrained hardware would allow imbuing many edge devices with intelligence. Microcontrollers (MCUs) are an attractive platform for building smart devices due to their low cost, wide availability, and modest power usage. However, they lack the computational resources to run neural networks as straightforwardly as mobile or server platforms, which necessitates changes to the network architecture and the inference software. In this work, we discuss the deployment and memory concerns of neural networks on MCUs and present a way of saving memory by changing the execution order of the network’s operators, which is orthogonal to other compression methods. We publish a tool for reordering operators of TensorFlow Lite models and demonstrate its utility by sufficiently reducing the memory footprint of a CNN to deploy it on an MCU with 512KB SRAM.

Materials

  • PDF
  • Code
BibTeX
@misc{liberis2019neural,
  title = {{Neural networks on microcontrollers: saving memory at inference via operator reordering}},
  author = {Liberis, Edgar and Lane, Nicholas D.},
  year = {2019},
  url = {https://arxiv.org/abs/1910.05110},
  eprint = {1910.05110},
  archiveprefix = {arXiv}
}
Cambridge Machine Learning Systems Group
Department of Computer Science and Technology
University of Cambridge
ContactGitHub