A Neural Network-Based On-Device Learning Anomaly Detector for Edge Devices
A Neural Network-Based On-Device Learning Anomaly Detector for Edge Devices
复制标题
DOI：
10.1109/tc.2020.2973631
复制
发表时间：
2020-07-01
影响因子：
3.7
通讯作者：
Matsutani, Hiroki
中科院分区：
文献类型：
--
作者：
Tsukada, Mineto;Kondo, Masaaki;Matsutani, Hiroki
半监督异常检测是一种通过学习正常数据的分布来识别异常的方法。反向传播神经网络（即，基于BP神经网络的方法由于其良好的泛化能力最近引起了人们的关注。在典型情况下，基于BP-NN的模型在服务器机器中使用从边缘设备收集的输入数据进行迭代优化。然而，（1）迭代优化通常需要显著的努力来跟随正态数据的分布的变化（即，概念漂移），以及（2）边缘和服务器之间的数据传输强加了额外的等待时间和能量消耗。为了解决这些问题，我们提出了ONLAD及其IP核，命名为ONLAD核心。ONLAD经过高度优化，可执行快速顺序学习，在不到1毫秒的时间内跟踪概念漂移。ONLAD Core以低功耗实现边缘设备的设备上学习，从而实现独立执行，无需在边缘和服务器之间传输数据。实验表明，ONLAD在模拟概念漂移的环境中具有良好的异常检测能力。对ONLAD Core的评估证实，训练延迟比其他软件实现快1.95倍（类似于6.58倍）。此外，在小型FPGA/CPU SoC平台PYNQ-Z1板上实现的ONLAD Core的运行时功耗比它们低5.0倍至25.4倍。
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 the 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 similar to 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 similar to 25.4x lower than them.