edoc

Learning Extremal Representations with Deep Archetypal Analysis

Keller, Sebastian Mathias and Samarin, Maxim and Arend Torres, Fabricio and Wieser, Mario and Roth, Volker. (2021) Learning Extremal Representations with Deep Archetypal Analysis. International Journal of Computer Vision, 129 (4). pp. 805-820.

[img]
Preview
PDF - Published Version
Available under License CC BY (Attribution).

2883Kb

Official URL: https://edoc.unibas.ch/81729/

Downloads: Statistics Overview

Abstract

Archetypes represent extreme manifestations of a population with respect to specific characteristic traits or features. In linear feature space, archetypes approximate the data convex hull allowing all data points to be expressed as convex mixtures of archetypes. As mixing of archetypes is performed directly on the input data, linear Archetypal Analysis requires additivity of the input, which is a strong assumption unlikely to hold e.g. in case of image data. To address this problem, we propose learning an appropriate latent feature space while simultaneously identifying suitable archetypes. We thus introduce a generative formulation of the linear archetype model, parameterized by neural networks. By introducing the distance-dependent archetype loss, the linear archetype model 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. Moreover, the information bottleneck framework allows for a natural incorporation of arbitrarily complex side information during training. As a consequence, learned archetypes become easily interpretable as they derive their meaning directly from the included side information. Applicability of the proposed method is demonstrated by exploring archetypes of female facial expressions while using multi-rater based emotion scores of these expressions as side information. A second application illustrates the exploration of the chemical space of small organic molecules. By using different kinds of side information we demonstrate how identified archetypes, along with their interpretation, largely depend on the side information provided.
Faculties and Departments:05 Faculty of Science > Departement Mathematik und Informatik > Informatik > Biomedical Data Analysis (Roth)
UniBasel Contributors:Roth, Volker and Keller, Sebastian Mathias and Samarin, Maxim and Arend Torres, Fabricio and Wieser, Mario
Item Type:Article, refereed
Article Subtype:Research Article
Publisher:Springer
ISSN:0920-5691
e-ISSN:1573-1405
Note:Publication type according to Uni Basel Research Database: Journal article
Language:English
Related URLs:
Identification Number:
edoc DOI:
Last Modified:07 Jun 2024 14:50
Deposited On:12 Apr 2021 12:34

Repository Staff Only: item control page