4
views
0
recommends
+1 Recommend
1 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      TEMPORAL DICTIONARY LEARNING FOR TIME-SERIES DECOMPOSITION Translated title: APRENDIZAJE DE DICCIONARIOS TEMPORALES PARA LA DESCOMPOSICIÓN DE SERIES DE TIEMPOS

      research-article

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Dictionary Learning (DL) is a feature learning method that derives a finite collection of dictionary elements (atoms) from a given dataset. These atoms are small characteristic features representing recurring patterns within the data. A dictionary therefore is a compact representation of complex or large scale datasets. In this paper we investigate DL for temporal signal decomposition and reconstruction. Decomposition is a common method in time-series forecasting to separate a complex composite signal into different frequency components as to reduce forecasting complexity. By representing characteristic features, we consider dictionary elements to function as filters for the decomposition of temporal signals. Rather than simple filters with clearly defined frequency spectra, we hypothesize for dictionaries and the corresponding reconstructions to act as more complex filters. Training different dictionaries then permits to decompose the original signal into different components. This makes it a potential alternative to existing decomposition methods. We apply a known sparse DL algorithm to a wind speed dataset and investigate decomposition quality and filtering characteristics. Reconstruction accuracy serves as a proxy for evaluating the dictionary quality and a coherence analysis is performed to analyze how different dictionary configurations lead to different filtering characteristics. The results of the presented work demonstrate how learned features of different dictionaries represent transfer functions corresponding to frequency components found in the original data. Based on finite sets of atoms, dictionaries provide a deterministic mechanism to decompose a signal into various reconstructions and their respective remainders. These insights have direct application to the investigation and development of advanced signal decomposition and forecasting techniques.

          Translated abstract

          Dictionary Learning (DL) es un método de aprendizaje de características que deriva una colección finita de elementos del diccionario (átomos) de un conjunto de datos determinado. Estos átomos son pequeños rasgos característicos que representan patrones recurrentes dentro de los datos. Por lo tanto, un diccionario es una representación compacta de conjuntos de datos complejos o de gran escala. En este trabajo investigamos DL para la descomposición y reconstrucción de señales temporales. La descomposición es un método común en el pronóstico de series de tiempo para separar una señal compuesta compleja en diferentes componentes de frecuencia para reducir la complejidad del pronóstico. Al representar los rasgos característicos, consideramos que los elementos del diccionario funcionan como filtros para la descomposición de las señales temporales. En lugar de filtros simples con espectros de frecuencia claramente definidos, planteamos la hipótesis de que los diccionarios y las reconstrucciones correspondientes actúen como filtros más complejos. La capacitación de diferentes diccionarios permite luego descomponer la señal original en diferentes componentes. Esto lo convierte en una alternativa potencial a los métodos de descomposición existentes. Aplicamos un algoritmo de DL disperso conocido a un conjunto de datos de velocidad del viento e investigamos la calidad de descomposición y las características de filtrado. La precisión de la reconstrucción sirve como un proxy para evaluar la calidad del diccionario y se realiza un análisis de coherencia para analizar cómo diferentes configuraciones de diccionarios llevan a diferentes características de filtrado. Los resultados del trabajo presentado demuestran cómo las características aprendidas de diferentes diccionarios representan funciones de transferencia correspondientes a los componentes de frecuencia encontrados en los datos originales. Basados en conjuntos finitos de átomos, los diccionarios proporcionan un mecanismo determinista para descomponer una señal en varias reconstrucciones y sus respectivos residuos. Estos conocimientos tienen una aplicación directa en la investigación y el desarrollo de técnicas avanzadas de descomposición de señales y pronóstico.

          Related collections

          Most cited references11

          • Record: found
          • Abstract: not found
          • Article: not found

          The NCEP Climate Forecast System Version 2

            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found

            Rapid task-related plasticity of spectrotemporal receptive fields in primary auditory cortex.

            We investigated the hypothesis that task performance can rapidly and adaptively reshape cortical receptive field properties in accord with specific task demands and salient sensory cues. We recorded neuronal responses in the primary auditory cortex of behaving ferrets that were trained to detect a target tone of any frequency. Cortical plasticity was quantified by measuring focal changes in each cell's spectrotemporal response field (STRF) in a series of passive and active behavioral conditions. STRF measurements were made simultaneously with task performance, providing multiple snapshots of the dynamic STRF during ongoing behavior. Attending to a specific target frequency during the detection task consistently induced localized facilitative changes in STRF shape, which were swift in onset. Such modulatory changes may enhance overall cortical responsiveness to the target tone and increase the likelihood of 'capturing' the attended target during the detection task. Some receptive field changes persisted for hours after the task was over and hence may contribute to long-term sensory memory.
              Bookmark
              • Record: found
              • Abstract: not found
              • Article: not found

              Learning Overcomplete Representations

                Bookmark

                Author and article information

                Journal
                riyd
                Investigación & Desarrollo
                Inv. y Des.
                UNIVERSIDAD PRIVADA BOLIVIANA (Cochabamba, , Bolivia )
                1814-6333
                2518-4431
                2019
                : 19
                : 1
                : 105-112
                Affiliations
                [01] orgnameUniversidad Privada Boliviana orgdiv1Institute for Computational Intelligence (ICI) jensburger@ 123456upb.edu
                Article
                S2518-44312019000100008 S2518-4431(19)01900100008
                871c406e-d909-476b-bc9e-715686210c9a

                This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

                History
                : 26 June 2019
                : 04 June 2019
                Page count
                Figures: 0, Tables: 0, Equations: 0, References: 14, Pages: 8
                Product

                SciELO Bolivia

                Categories
                ARTICLES - ENGINEERING

                SAILnet,Series de Tiempo,Dictionary Learning,Decomposition,Time-Series,Descomposición

                Comments

                Comment on this article