mlsanityData to 7 Oct 2026

On-Device Training of Fully Quantized Deep Neural Networks on Cortex-M Microcontrollers

arXiv 2407.10734 ↗First posted 15 Jul 2024cs.LG cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

On-device training of DNNs allows models to adapt and fine-tune to newly collected data or changing domains while deployed on microcontroller units (MCUs). However, DNN training is a resource-intensive task, making the implementation and execution of DNN training algorithms on MCUs challenging due to low processor speeds, constrained throughput, limited floating-point support, and memory constraints. In this work, we explore on-device training of DNNs for Cortex-M MCUs. We present a method that enables efficient training of DNNs completely in place on the MCU using fully quantized training (FQT) and dynamic partial gradient updates. We demonstrate the feasibility of our approach on multiple vision and time-series datasets and provide insights into the tradeoff between training accuracy, memory overhead, energy, and latency on real hardware.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 23 of its 45 references that are in the map. First posted on arXiv 15 Jul 2024.

Builds on

23 of 45 references.

References · 23

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  3. 2019Searching for MobileNetV3cited by 1,984
  4. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
  5. 2016Training Deep Nets with Sublinear Memory Costcited by 825
  6. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
  7. 2018Quantizing deep convolutional networks for efficient inference: A whitepapercited by 507
  8. 2018Deep Learning for Classical Japanese Literaturecited by 415
  9. 2020MCUNet: Tiny Deep Learning on IoT Devicescited by 269
  10. 2018Training Deep Neural Networks with 8-bit Floating Point Numberscited by 163
  11. 2022On-Device Training Under 256KB Memorycited by 158
  12. 2020Training BatchNorm and Only BatchNorm: On the Expressive Power of Random Features in CNNscited by 103
11 more
  1. 2016Memory-Efficient Backpropagation Through Timecited by 78
  2. 2021TinyOL: TinyML with Online-Learning on Microcontrollerscited by 65
  3. 2019Visual Wake Words Datasetcited by 46
  4. 2024Tiny Machine Learning: Progress and Futurescited by 40
  5. 2019E2-Train: Training State-of-the-art CNNs with Over 80% Energy Savingscited by 36
  6. 2018K for the Price of 1: Parameter-efficient Multi-task and Transfer Learningcited by 34
  7. 2022Mandheling: Mixed-Precision On-Device DNN Training with DSP Offloadingcited by 29
  8. 2022POET: Training Neural Networks on Tiny Devices with Integrated Rematerialization and Pagingcited by 26
  9. 2018Dynamic Sparse Graph for Efficient Deep Learningcited by 18
  10. 2022Energy-efficient Deployment of Deep Learning Applications on Cortex-M based Microcontrollers using Deep Compressioncited by 6
  11. 2024OpTC -- A Toolchain for Deployment of Neural Networks on AURIX TC3xx Microcontrollerscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 4.

5 papers.

  1. 2025Neural Network Quantization for Microcontrollers: A Comprehensive Survey of Methods, Platforms, and Applicationscited by 6
  2. 2026What changes after deployment? A survey on On-device Learning in TinyMLcited by 0
  3. 2026Hardware-Software Co-Design for Float16 On-Device Training on RISC-V Single-Corecited by 0
  4. 2026Design-Time Optimization of Deep Neural Networks for Intermittent Learning on Microcontrollerscited by 0
  5. 2026FORGE: Forward-Only Test-Time Adaptation for Integer-Only Vision Models on Microcontrollerscited by 0