Probabilistic archetypal analysis

  title={Probabilistic archetypal analysis},
  author={Sohan Seth and Manuel J. A. Eugster},
  journal={Machine Learning},
Archetypal analysis represents a set of observations as convex combinations of pure patterns, or archetypes. The original geometric formulation of finding archetypes by approximating the convex hull of the observations assumes them to be real–valued. This, unfortunately, is not compatible with many practical situations. In this paper we revisit archetypal analysis from the basic principles, and propose a probabilistic framework that accommodates other observation types such as integers, binary… 

Archetypal Analysis for Nominal Observations

  • S. SethM. Eugster
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2016
This article views archetypal analysis in a generative framework: this allows explicit control over choosing a suitable number of archetypes by assigning appropriate prior information, and finding efficient update rules using variational Bayes'.

Archetypal shapes based on landmarks and extension to handle missing data

This work proposes new algorithms for performing archetypal analysis with missing data and its extension to incomplete shapes, and proposes new tools that lie somewhere between clustering and matrix factorization methods.

A Geometric Approach to Archetypal Analysis via Sparse Projections

Experimental results show that GAA is considerably faster while performing comparable to existing methods for tasks such as classification, data visualization/categorization.

Archetypal analysis for ordinal data

Consistency of Archetypal Analysis

This paper proves a consistency result that shows if the data is independently sampled from a probability measure with bounded support, then the archetype points converge to a solution of the continuum version of the problem, of which it identifies and establishes several properties.

Finding Archetypal Spaces Using Neural Networks

AAnet is introduced, which is a deep neural network framework for learning and generating from a latent archetypal representation of data, and it is demonstrated that AAnet can generate from data geometry rather than from data density, and is used to identify biologically meaningful archetypes in single-cell gene expression data.

Deep Archetypal Analysis

"Deep Archetypal Analysis" generates latent representations of high-dimensional datasets in terms of fractions of intuitively understandable basic entities called archetypes, an unsupervised method to represent multivariate data points as sparse convex combinations of extremal elements of the dataset.

Learning Extremal Representations with Deep Archetypal Analysis

This work introduces a generative formulation of the linear archetype model, parameterized by neural networks, which can be integrated into the latent space of a deep variational information bottleneck and an optimal representation, together with the archetypes, can be learned end-to-end.

Robust multivariate and functional archetypal analysis with application to financial time series analysis

Archetypal Analysis With Missing Data: See All Samples by Looking at a Few Based on Extreme Profiles

This article proposes several methodologies for handling missing or incomplete data in archetype analysis (AA) and archetypoid analysis (ADA) and illustrates how these unsupervised methodologies allow complex data to be understood, even by nonexperts.



Weighted and robust archetypal analysis

Making Archetypal Analysis Practical

This paper introduces ways of notably accelerating archetypal analysis and introduces the first successful application of the technique to large scale data analysis problems.

Archetypal analysis of spatio-temporal dynamics

Archetypal analysis for machine learning and data mining

Archetypal analysis of galaxy spectra

ABSTRACT Archetypal analysis represents each individual member of a set of data vectors as a mixture (aconstrained linear combination)of the pure types or archetypes of the data set. The

Archetypal Scientists

Face Recognition via Archetype Hull Ranking

A sensible face similarity measure is accomplished which can support robust and effective face recognition and verification together and is evaluated in terms of experiments performed on three benchmark face databases Multi-PIE, Pubfig83, and LFW.

On the use of archetypes as benchmarks

A three-step data driven benchmarking procedure is suggested, which enables users to identify some reference performers, to analyze their features, and to compare observed performers with them.

Hierarchical Convex NMF for Clustering Massive Data

An extension of convex-hull non-negative matrix factorization (CH-NMF) that automatically adapts to internal structures of a dataset, hence it yields meaningful and interpretable clusters for non-convex datasets.

Performance Profiles based on Archetypal Athletes

Performance indicators and, on their basis, performance profiles are one of the foundations of performance analysis in sports. Obviously, the crux is to develop performance profiles which allow to