Robust Online Time Series Prediction with Recurrent Neural Networks
Authors:
- Tian GuoTGTian Guo
- Zhao XuZXZhao Xu
- Xin YaoXYXin Yao
- Haifeng ChenHCHaifeng Chen
- Karl AbererKAKarl Aberer
- Koichi FunayaKFKoichi Funaya
- 1 October 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 816-825
Abstract
Time series forecasting for streaming data plays an important role in many real applications, ranging from IoT systems, cyber-networks, to industrial systems and healthcare. However the real data is often complicated with anomalies and change points, which can lead the learned models deviating from the underlying patterns of the time series, especially in the context of online learning mode. In this paper we present an adaptive gradient learning method for recurrent neural networks (RNN) to forecast streaming time series in the presence of anomalies and change points. We explore the local features of time series to automatically weight the gradients of the loss of the newly available observations with distributional properties of the data in real time. We perform extensive experimental analysis on both synthetic and real datasets to evaluate the performance of the proposed method.Keywords
This publication has 17 references indexed in Scilit:
- Anomaly detection in ECG time signals via deep long short-term memory networks2015
- Outlier Detection for Temporal DataSynthesis Lectures on Data Mining and Knowledge Discovery, 2014
- A survey on concept drift adaptationACM Computing Surveys, 2014
- A time-dependent enhanced support vector machine for time series regression2013
- An empirical study of learning rates in deep neural networks for speech recognition2013
- DDD: A New Ensemble Approach for Dealing with Concept DriftIEEE Transactions on Knowledge and Data Engineering, 2011
- Sequential Bayesian prediction in the presence of changepoints2009
- Learning to Forget: Continual Prediction with LSTMNeural Computation, 2000
- Long Short-Term MemoryNeural Computation, 1997
- Recurrent neural networks and robust time series predictionIEEE Transactions on Neural Networks, 1994