Non-Parametric Time Series Classification

We present an improved state-based prediction algorithm for time series. Given time series produced by a process composed of different underlying states, the algorithm predicts future time series values based on past time series values for each state. Unlike many algorithms, this algorithm predicts a multi-modal distribution over future values. This prediction forms the basis for labelling part of a time series with the underlying state that created it given some labelled example signals. The algorithm is robust to a wide variety of possible types of changes in signals including changes in mean, amplitude, amount of noise, and period. We show results demonstrating that the algorithm successfully segments signals from several robotic sensors generated while performing a variety of simple tasks.

[1]  Robert P. W. Duin,et al.  On the Choice of Smoothing Parameters for Parzen Estimators of Probability Density Functions , 1976, IEEE Transactions on Computers.

[2]  Athanasios Kehagias,et al.  Modular neural networks for MAP classification of time series and the partition algorithm , 1996, IEEE Trans. Neural Networks.

[3]  Manuela M. Veloso,et al.  Classification of robotic sensor streams using non-parametric statistics , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[4]  Manuela M. Veloso,et al.  Automatic detection and response to environmental change , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).

[5]  Vassilios Petridis,et al.  Predictive Modular Neural Networks: Applications to Time Series , 1998 .

[6]  Andrew W. Moore,et al.  Learning to recognize time series: combining ARMA models with memory-based learning , 1997, Proceedings 1997 IEEE International Symposium on Computational Intelligence in Robotics and Automation CIRA'97. 'Towards New Computational Principles for Robotics and Automation'.

[7]  Geoffrey E. Hinton,et al.  Switching State-Space Models , 1996 .

[8]  William D. Penny,et al.  Dynamic Models for Nonstationary Signal Segmentation , 1999, Comput. Biomed. Res..

[9]  Michèle Basseville,et al.  Detection of abrupt changes: theory and application , 1993 .

[10]  Geoffrey E. Hinton,et al.  Variational Learning for Switching State-Space Models , 2000, Neural Computation.

[11]  K. Obermayer,et al.  Multiple-step ahead prediction for non linear dynamic systems: A Gaussian Process treatment with propagation of the uncertainty , 2003, NIPS 2003.

[12]  Masafumi Hashimoto,et al.  Sensor fault detection and identification in dead-reckoning system of mobile robot: interacting multiple model approach , 2001, Proceedings 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems. Expanding the Societal Role of Robotics in the the Next Millennium (Cat. No.01CH37180).