Archetypal Analysis of Cellular Flame Data
نویسندگان
چکیده
The application of archetypal analysis to high-dimensional data arising from video-taped images is presented. A hybrid principal components/archetypes technique has been developed to overcome the difficulties of applying archetypes to data sets with points living in a space of dimension higher than about 500. The advantages of the method lie in the creation of patterns typical of the set as a whole, and an expression of the dynamics in terms of these patterns.
منابع مشابه
A Note on Archetypal Analysis and the Approximation of Convex Hulls
We briefly review the basic ideas behind archetypal analysis for matrix factorization and discuss its behavior in approximating the convex hull of a data sample. We then ask how good such approximations can be and consider different cases. Understanding archetypal analysis as the problem of computing a convexity constrained low-rank approximation of the identity matrix provides estimates for ar...
متن کاملMaking Archetypal Analysis Practical
Archetypal analysis represents the members of a set of multivariate data as a convex combination of extremal points of the data. It allows for dimensionality reduction and clustering and is particularly useful whenever the data are superpositions of basic entities. However, since its computation costs grow quadratically with the number of data points, the original algorithm hardly applies to mo...
متن کاملCharacterization of Freely Propagating Hydrogen Flames
We explore the detailed structure of heat release and the energy budget for a freely-propagating lean, premixed hydrogen flame. At lean conditions, the flame is thermodiffusively unstable and spontaneously develops the cellular burning structures, characteristic of this type of flame. The instabilities in the system saturate quickly, resulting in robust, slowly evolving cellular burning feature...
متن کاملWeighted and robust archetypal analysis
Archetypal analysis represents observations in a multivariate data set as convex combinations of a few extremal points lying on the boundary of the convex hull. Data points which vary from the majority have great influence on the solution; in fact one outlier can break down the archetype solution. This paper adapts the original algorithm to be a robust M-estimator and presents an iteratively re...
متن کاملArchetypal Analysis of Interval Data
In this paper we present a mathematical model for archetypal analysis of data represented by means of intervals of real numbers. We extend the model for single-valued data proposed in the pioneering work of Cutler and Breiman on this topic. The core problem is a non-convex optimization one, which we solve by means of a sequential quadratic programming method. We show numerical experiments perfo...
متن کامل