Satellite Anomaly Detection Using Variance Based Genetic Ensemble of Neural Networks

被引:1
作者
Sadr, Mohammad Amin Maleki [1 ]
Zhu, Yeying [1 ]
Hu, Peng [1 ,2 ]
机构
[1] Univ Waterloo, Waterloo, ON N2L 3G1, Canada
[2] Natl Res Council Canada, Waterloo, ON N2L 3G1, Canada
来源
ICC 2023-IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS | 2023年
关键词
Anomaly detection; Genetic Algorithm; Neural Networks; LSTM; RNN; GRU;
D O I
10.1109/ICC45041.2023.10278933
中图分类号
TN [电子技术、通信技术];
学科分类号
0809 ;
摘要
In this paper, we use a variance-based genetic ensemble (VGE) of Neural Networks (NNs) to detect anomalies in the satellite's historical data. We use an efficient ensemble of the predictions from multiple Recurrent Neural Networks (RNNs) by leveraging each model's uncertainty level (variance). For prediction, each RNN is guided by a Genetic Algorithm (GA) which constructs the optimal structure for each RNN model. However, finding the model uncertainty level is challenging in many cases. Although the Bayesian NNs (BNNs)-based methods are popular for providing the confidence bound of the models, they cannot be employed in complex NN structures as they are computationally intractable. This paper uses the Monte Carlo (MC) dropout as an approximation version of BNNs. Then these uncertainty levels and each predictive model suggested by GA are used to generate a new model, which is then used for forecasting the TS and AD. Simulation results show that the forecasting and AD capability of the ensemble model outperforms existing approaches.
引用
收藏
页码:4070 / 4075
页数:6
相关论文
共 20 条
[1]   LOF: Identifying density-based local outliers [J].
Breunig, MM ;
Kriegel, HP ;
Ng, RT ;
Sander, J .
SIGMOD RECORD, 2000, 29 (02) :93-104
[2]   An enhanced fast non-dominated solution sorting genetic algorithm for multi-objective problems [J].
Deng, Wu ;
Zhang, Xiaoxiao ;
Zhou, Yongquan ;
Liu, Yi ;
Zhou, Xiangbing ;
Chen, Huiling ;
Zhao, Huimin .
INFORMATION SCIENCES, 2022, 585 :441-453
[3]   Ensemble Neuroevolution-Based Approach for Multivariate Time Series Anomaly Detection [J].
Faber, Kamil ;
Pietron, Marcin ;
Zurek, Dominik .
ENTROPY, 2021, 23 (11)
[4]  
Gal Y, 2016, ADV NEUR IN, V29
[5]  
Gal Y, 2016, PR MACH LEARN RES, V48
[6]   TadGAN: Time Series Anomaly Detection Using Generative Adversarial Networks [J].
Geiger, Alexander ;
Liu, Dongyu ;
Alnegheimish, Sarah ;
Cuesta-Infante, Alfredo ;
Veeramachaneni, Kalyan .
2020 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2020, :33-43
[7]   Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding [J].
Hundman, Kyle ;
Constantinou, Valentino ;
Laporte, Christopher ;
Colwell, Ian ;
Soderstrom, Tom .
KDD'18: PROCEEDINGS OF THE 24TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2018, :387-395
[8]   MAD-GAN: Multivariate Anomaly Detection for Time Series Data with Generative Adversarial Networks [J].
Li, Dan ;
Chen, Dacheng ;
Shi, Lei ;
Jin, Baihong ;
Goh, Jonathan ;
Ng, See-Kiong .
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2019: TEXT AND TIME SERIES, PT IV, 2019, 11730 :703-716
[9]   Intelligent Analysis Algorithm for Satellite Health under Time-Varying and Extremely High Thermal Loads [J].
Li, En-Hui ;
Li, Yun-Ze ;
Li, Tian-Tian ;
Li, Jia-Xin ;
Zhai, Zhuang-Zhuang ;
Li, Tong .
ENTROPY, 2019, 21 (10)
[10]   A General Framework for Uncertainty Estimation in Deep Learning [J].
Loquercio, Antonio ;
Segu, Mattia ;
Scaramuzza, Davide .
IEEE ROBOTICS AND AUTOMATION LETTERS, 2020, 5 (02) :3153-3160