Time Series Clustering with Deep Reservoir Computing
| Autor: | Alessio Micheli, Gonzalo Joya, Claudio Gallicchio, Miguel Atencia |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Structure (mathematical logic)
Time series Series (mathematics) Computer science Clustering Echo State Networks Reservoir Computing Reservoir computing 02 engineering and technology computer.software_genre Cluster 020204 information systems 0202 electrical engineering electronic engineering information engineering Benchmark (computing) Cluster (physics) 020201 artificial intelligence & image processing Data mining Cluster analysis computer Análisis cluster -- Programas de ordenador |
| Zdroj: | Artificial Neural Networks and Machine Learning – ICANN 2020 ISBN: 9783030616151 ICANN (2) RIUMA: Repositorio Institucional de la Universidad de Málaga Universidad de Málaga RIUMA. Repositorio Institucional de la Universidad de Málaga instname |
| Popis: | This paper proposes a method for clustering of time series, based upon the ability of deep Reservoir Computing networks to grasp the dynamical structure of the series that is presented as input. A standard clustering algorithm, such as k-means, is applied to the network states, rather than the input series themselves. Clustering is thus embedded into the network dynamical evolution, since a clustering result is obtained at every time step, which in turn serves as initialisation at the next step. We empirically assess the performance of deep reservoir systems in time series clustering on benchmark datasets, considering the influence of crucial hyperparameters. Experimentation with the proposed model shows enhanced clustering quality, measured by the silhouette coefficient, when compared to both static clustering of data, and dynamic clustering with a shallow network. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|