Hierarchical Linear Dynamical Systems: A new model for clustering of time series

Goktug T. Cinar, Carlos A. Loza, Jose C. Principe

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

5 Citas (Scopus)

Resumen

The auditory cortex in the brain does effortlessly a better job of extracting information from the acoustic world than our current generation of signal processing algorithms. The proposed architecture, Hierarchical Linear Dynamical System (HLDS), is based on Kaiman filters with hierarchically coupled state models that stabilize the input dynamics and provide a representation space. This approach extracts information from the input and self-organizes it in the higher layers leading to an algorithm capable of clustering time series in an unsupervised manner. In this paper we further investigate the properties of HLDS, demonstrate its performance on music rather than isolated notes and propose the time domain implementation to overcome one of its current bottlenecks.

Idioma originalInglés
Título de la publicación alojadaProceedings of the International Joint Conference on Neural Networks
EditorialInstitute of Electrical and Electronics Engineers Inc.
Páginas2464-2470
Número de páginas7
ISBN (versión digital)9781479914845
DOI
EstadoPublicada - 3 sep. 2014
Publicado de forma externa
Evento2014 International Joint Conference on Neural Networks, IJCNN 2014 - Beijing, China
Duración: 6 jul. 201411 jul. 2014

Serie de la publicación

NombreProceedings of the International Joint Conference on Neural Networks

Conferencia

Conferencia2014 International Joint Conference on Neural Networks, IJCNN 2014
País/TerritorioChina
CiudadBeijing
Período6/07/1411/07/14

Huella

Profundice en los temas de investigación de 'Hierarchical Linear Dynamical Systems: A new model for clustering of time series'. En conjunto forman una huella única.

Citar esto