# Decentor-V: Lightweight ML Training on Low-Power RISC-V Edge Devices

arXiv: https://arxiv.org/abs/2509.18118 · HTML: https://mlsanity.com/paper/2509.18118/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 33 references that are in the map.
- First posted on arXiv 10 Sep 2025.

## Builds on (most cited first, 13 of 13)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Federated Learning: Challenges, Methods, and Future Directions](https://mlsanity.com/paper/1908.07873/) (2019)
- [An overview of gradient descent optimization algorithms](https://mlsanity.com/paper/1609.04747/) (2016)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [CMSIS-NN: Efficient Neural Network Kernels for Arm Cortex-M CPUs](https://mlsanity.com/paper/1801.06601/) (2018)
- [Lightweight Deep Learning for Resource-Constrained Environments: A Survey](https://mlsanity.com/paper/2404.07236/) (2024)
- [PULP-NN: Accelerating Quantized Neural Networks on Parallel Ultra-Low-Power RISC-V Processors](https://mlsanity.com/paper/1908.11263/) (2019)
- [Federated learning: Applications, challenges and future directions](https://mlsanity.com/paper/2205.09513/) (2022)
- [Training Machine Learning models at the Edge: A Survey](https://mlsanity.com/paper/2403.02619/) (2024)
- [Performance Analysis and Comparison of Machine and Deep Learning Algorithms for IoT Data Classification](https://mlsanity.com/paper/2001.09636/) (2020)
- [Gradient Descent, Stochastic Optimization, and Other Tales](https://mlsanity.com/paper/2205.00832/) (2022)
- [Shifting Capsule Networks from the Cloud to the Deep Edge](https://mlsanity.com/paper/2110.02911/) (2021)
- [Learn What You Need in Personalized Federated Learning](https://mlsanity.com/paper/2401.08327/) (2024)

## Built on by (most cited first, 1 of 1)

- [Hardware-Software Co-Design for Float16 On-Device Training on RISC-V Single-Core](https://mlsanity.com/paper/2607.21130/) (2026)

## Abstract (arXiv)

Modern IoT devices increasingly rely on machine learning solutions to process data locally. However, the lack of graphics processing units (GPUs) or dedicated accelerators on most platforms makes on-device training largely infeasible, often requiring cloud-based services to perform this task. This procedure often raises privacy-related concerns, and creates dependency on reliable and always-on connectivity. Federated Learning (FL) is a new trend that addresses these issues by enabling decentralized and collaborative training directly on devices, but it requires highly efficient optimization algorithms. L-SGD, a lightweight variant of stochastic gradient descent, has enabled neural network training on Arm Cortex-M Microcontroller Units (MCUs). This work extends L-SGD to RISC-V-based MCUs, an open and emerging architecture that still lacks robust support for on-device training. L-SGD was evaluated on both Arm and RISC-V platforms using 32-bit floating-point arithmetic, highlighting the performance impact of the absence of Floating-Point Units (FPUs) in RISC-V MCUs. To mitigate these limitations, we introduce an 8-bit quantized version of L-SGD for RISC-V, which achieves nearly 4x reduction in memory usage and a 2.2x speedup in training time, with negligible accuracy degradation.