An Information Geometrical View of Stationary Subspace Analysis

Autor:	Paul von Bünau, Wojciech Samek, Motoaki Kawanabe, Frank C. Meinecke
Rok vydání:	2011
Předmět:	Kullback–Leibler divergence business.industry Pattern recognition Linear subspace Independent component analysis Generative model Stationary subspace analysis Unsupervised learning Information geometry Artificial intelligence Invariant (mathematics) business Algorithm Mathematics
Zdroj:	Lecture Notes in Computer Science ISBN: 9783642217371 ICANN (2)
DOI:	10.1007/978-3-642-21738-8_51
Popis:	Stationary Subspace Analysis (SSA) is an unsupervised learning method that finds subspaces in which data distributions stay invariant over time. It has been shown to be very useful for studying non-stationarities in various applications. In this paper, we present the first SSA algorithm based on a full generative model of the data. This new derivation relates SSA to previous work on finding interesting subspaces from high-dimensional data in a similar way as the three easy routes to independent component analysis, and provides an information geometric view.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::420799b418647bfd4c7db674cdeb587f https://doi.org/10.1007/978-3-642-21738-8_51 Zobrazit plný text záznamu