.png){kind=icon}
返回
In-situ animal behavior classification using knowledge distillation and fixed-point quantization
.png){kind=icon}
OA
AI
DOI：10.1016/j.atech.2022.100159.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
We explore the use of knowledge distillation (KD) for learning compact and accurate models that enable classification of animal behavior from accelerometry data on wearable devices. To this end, we take a deep and complex convolutional neural network, known as residual neural network (ResNet), as the teacher model. ResNet is specifically designed for multivariate time-series classification. We use ResNet to distill the knowledge of animal behavior classification datasets into soft labels, which consist of the predicted pseudo-probabilities of every class for each datapoint. We then use the soft labels to train our significantly less complex student models, which are based on the gated recurrent unit (GRU) and multilayer perceptron (MLP). The evaluation results using two real-world animal behavior classification datasets show that the classification accuracy of the student GRUMLP models improves appreciably through KD, approaching that of the teacher ResNet model. To further reduce the computational and memory requirements of performing inference using the student models trained via KD, we utilize dynamic fixed-point quantization (DQ) through an appropriate modification of the computational graph of the considered models. We implement both unquantized and quantized versions of the developed KDbased models on the embedded systems of our purpose-built collar and ear tag devices to classify animal behavior in situ and in real time. Our evaluations corroborate the effectiveness of KD and DQ in improving the accuracy and efficiency of in-situ animal behavior classification.
Keyword:
Animal behavior classification
Deep learning
Dynamic quantization
Embedded systems
Fix-point arithmetic
Knowledge distillation
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
5.7
论文数:
2.6K
被引数:
2.5K
机构
引用论文
.png){kind=icon}
SBA-15:TiO2 nanocomposites: II. Direct and post-synthesis using acetylacetoneSBA-15:TiO2纳米复合材料: II。使用乙酰丙酮的直接和后合成
PREAI
分享收藏Time-Series Classification with COTE: The Collective of Transformation-Based EnsemblesCOTE的时间序列分类: 基于变换的集合OAAI
OAAI分享收藏Predicting livestock behaviour using accelerometers: A systematic review of processing techniques for ruminant behaviour prediction from raw accelerometer data使用加速度计预测牲畜行为: 从原始加速度计数据预测反刍动物行为的处理技术的系统综述OAAI
OAAI分享收藏
查看更多内容