# A Neural Network-Based On-device Learning Anomaly Detector for Edge Devices

arXiv: https://arxiv.org/abs/1907.10147 · HTML: https://mlsanity.com/paper/1907.10147/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Jul 2019.

## Built on by (most cited first, 18 of 18)

- [Training Machine Learning models at the Edge: A Survey](https://mlsanity.com/paper/2403.02619/) (2024)
- [Addressing Gap between Training Data and Deployed Environment by On-Device Learning](https://mlsanity.com/paper/2203.01077/) (2022)
- [Generative Optimization Networks for Memory Efficient Data Generation](https://mlsanity.com/paper/2110.02912/) (2021)
- [A review of Federated Learning in Intrusion Detection Systems for IoT](https://mlsanity.com/paper/2204.12443/) (2022)
- [Machine Learning for the Control and Monitoring of Electric Machine Drives: Advances and Trends](https://mlsanity.com/paper/2110.05403/) (2021)
- [Fast fluorescence lifetime imaging analysis via extreme learning machine](https://mlsanity.com/paper/2203.13754/) (2022)
- [Achieving Pareto Optimality using Efficient Parameter Reduction for DNNs in Resource-Constrained Edge Environment](https://mlsanity.com/paper/2403.10569/) (2024)
- [A Tiny Supervised ODL Core with Auto Data Pruning for Human Activity Recognition](https://mlsanity.com/paper/2408.01283/) (2024)
- [Skip2-LoRA: A Lightweight On-device DNN Fine-tuning Method for Low-cost Edge Devices](https://mlsanity.com/paper/2410.21073/) (2024)
- [Training Multi-Layer Binary Neural Networks With Local Binary Error Signals](https://mlsanity.com/paper/2412.00119/) (2024)
- [An Overflow/Underflow-Free Fixed-Point Bit-Width Optimization Method for OS-ELM Digital Circuit](https://mlsanity.com/paper/2103.09791/) (2021)
- [A Sequential Concept Drift Detection Method for On-Device Learning on Low-End Edge Devices](https://mlsanity.com/paper/2212.09637/) (2022)
- [LightESD: Fully-Automated and Lightweight Anomaly Detection Framework for Edge Computing](https://mlsanity.com/paper/2305.12266/) (2023)
- [An FPGA-Based Accelerator for Graph Embedding using Sequential Training Algorithm](https://mlsanity.com/paper/2312.15138/) (2023)
- [Developing a Resource-Constraint EdgeAI model for Surface Defect Detection](https://mlsanity.com/paper/2401.05355/) (2023)
- [Towards On-Device Learning and Reconfigurable Hardware Implementation for Encoded Single-Photon Signal Processing](https://mlsanity.com/paper/2504.09028/) (2025)
- [InstantFT: An FPGA-Based Runtime Subsecond Fine-tuning of CNN Models](https://mlsanity.com/paper/2506.06505/) (2025)
- [Semi-Supervised Online Learning on the Edge by Transforming Knowledge from Teacher Models](https://mlsanity.com/paper/2512.16866/) (2025)

## Abstract (arXiv)

Semi-supervised anomaly detection is an approach to identify anomalies by learning the distribution of normal data. Backpropagation neural networks (i.e., BP-NNs) based approaches have recently drawn attention because of their good generalization capability. In a typical situation, BP-NN-based models are iteratively optimized in server machines with input data gathered from edge devices. However, (1) the iterative optimization often requires significant efforts to follow changes in the distribution of normal data (i.e., concept drift), and (2) data transfers between edge and server impose additional latency and energy consumption. To address these issues, we propose ONLAD and its IP core, named ONLAD Core. ONLAD is highly optimized to perform fast sequential learning to follow concept drift in less than one millisecond. ONLAD Core realizes on-device learning for edge devices at low power consumption, which realizes standalone execution where data transfers between edge and server are not required. Experiments show that ONLAD has favorable anomaly detection capability in an environment that simulates concept drift. Evaluations of ONLAD Core confirm that the training latency is 1.95x~6.58x faster than the other software implementations. Also, the runtime power consumption of ONLAD Core implemented on PYNQ-Z1 board, a small FPGA/CPU SoC platform, is 5.0x~25.4x lower than them.