Independent Component Analysis (ICA) can be described in several ways, one of which is as a technique that seeks to find a set directions (components) underlying multivariate data that are most independent of one another. While there are several ICA models and many ICA methods, in this report we focus on the most basic model and one of the most popular and simple algorithms; the One-Unit FastIC...

Blind separation of signals through the info-max algorithm may be viewed as maximum likelihood learning in a latent variable model. In this paper we present an alternative approach to maximum likelihood learning in these models, namely Bayesian inference. It has already been shown how Bayesian inference can be applied to determine latent dimensionality in principal component analysis models (Bi...

Deriving resource properties of software components is a prerequisite for the specification of non-functional properties allowing a developer to also consider these properties for the application to build. In this paper, we describe a platform-independent approach to measure Java components in terms of a selected set of basic constituents (so called atoms), which partition a program and are use...

Independent component analysis (ICA) aims at extracting unknown hidden factors/components from multivariate data using only the assumption that the unknown factors are mutually independent. Since the introduction of ICA concepts in the early 1980s in the context of neural networks and array signal processing, many new successful algorithms have been proposed that are now well-established method...

to Michaela iv Preface If we knew what we were doing, it wouldn't be called research, would it? 1 Imagine you are a robot. Imagine you are a robot visiting a cocktail party — groups of people standing around chatting with each other. Imagine you are a robot trying to understand what those people talk about. You have no problems when facing one of those humans alone. You have built-in text-to-sp...

Accurately evaluating statistical independence among random variables is a key element of independent component analysis (ICA). In this letter, we employ a squared-loss variant of mutual information as an independence measure and give its estimation method. Our basic idea is to estimate the ratio of probability densities directly without going through density estimation, thereby avoiding the di...

Linear Independent Component Analysis (ICA) has become an important technique in unsupervised neural learning. Even though linear ICA yields meaningful results in many cases, it can provide a crude approximation only for general nonlinear data distributions. In this paper we study techniques where local ICA models are applied to data rst grouped or clustered using some suitable algorithm. The g...

Independent component analysis is a probabilistic method for learning a linear transform of a random vector. The goal is to find components that are maximally independent and non-Gaussian (non-normal). Its fundamental difference to classical multi-variate statistical methods is in the assumption of non-Gaussianity, which enables the identification of original, underlying components, in contrast...

A common method for texture representation is to use the marginal probability densities over the outputs of a set of multi-orientation, multi-scale filters as a description of the texture. We propose a technique, based on Independent Components Analysis, for choosing the set of filters that yield the most informative marginals, meaning that the product over the marginals most closely approximat...