# Training on the Edge: The why and the how

arXiv: https://arxiv.org/abs/1903.03051 · HTML: https://mlsanity.com/paper/1903.03051/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Feb 2019.

## Built on by (most cited first, 9 of 9)

- [TinyOL: TinyML with Online-Learning on Microcontrollers](https://mlsanity.com/paper/2103.08295/) (2021)
- [Distributed intelligence on the Edge-to-Cloud Continuum: A systematic literature review](https://mlsanity.com/paper/2205.01081/) (2022)
- [ECAvg: An Edge-Cloud Collaborative Learning Approach using Averaged Weights](https://mlsanity.com/paper/2310.03823/) (2023)
- [Achieving Pareto Optimality using Efficient Parameter Reduction for DNNs in Resource-Constrained Edge Environment](https://mlsanity.com/paper/2403.10569/) (2024)
- [Regularization-based Framework for Quantization-, Fault- and Variability-Aware Training](https://mlsanity.com/paper/2503.01297/) (2025)
- [Performance Analysis of DNN Inference/Training with Convolution and non-Convolution Operations](https://mlsanity.com/paper/2306.16767/) (2023)
- [Developing a Resource-Constraint EdgeAI model for Surface Defect Detection](https://mlsanity.com/paper/2401.05355/) (2023)
- [Online Continual Learning on Intel Loihi 2 via a Co-designed Spiking Neural Network](https://mlsanity.com/paper/2511.01553/) (2025)
- [Breaking Chains with Trees: Model-Parallel Deep Learning with O(log N) Time Complexity](https://mlsanity.com/paper/2606.21497/) (2026)

## Abstract (arXiv)

Edge computing is the natural progression from Cloud computing, where, instead of collecting all data and processing it centrally, like in a cloud computing environment, we distribute the computing power and try to do as much processing as possible, close to the source of the data. There are various reasons this model is being adopted quickly, including privacy, and reduced power and bandwidth requirements on the Edge nodes. While it is common to see inference being done on Edge nodes today, it is much less common to do training on the Edge. The reasons for this range from computational limitations, to it not being advantageous in reducing communications between the Edge nodes. In this paper, we explore some scenarios where it is advantageous to do training on the Edge, as well as the use of checkpointing strategies to save memory.