Deep Archetypal Analysis

Keller, Sebastian Mathias and Samarin, Maxim and Wieser, Mario and Roth, Volker. (2019) Deep Archetypal Analysis. In: Pattern Recognition. Cham, pp. 171-185.

Full text not available from this repository.

Official URL: https://edoc.unibas.ch/73938/

Downloads: Statistics Overview


Deep Archetypal Analysis (DeepAA) generates latent representations of high-dimensional datasets in terms of intuitively understandable basic entities called archetypes. The proposed method extends linear Archetypal Analysis (AA), an unsupervised method to represent multivariate data points as convex combinations of extremal data points. Unlike the original formulation, Deep AA is generative and capable of handling side information. In addition, our model provides the ability for data-driven representation learning which reduces the dependence on expert knowledge. We empirically demonstrate the applicability of our approach by exploring the chemical space of small organic molecules. In doing so, we employ the archetype constraint to learn two different latent archetype representations for the same dataset, with respect to two chemical properties. This type of supervised exploration marks a distinct starting point and let us steer de novo molecular design.
Faculties and Departments:05 Faculty of Science > Departement Mathematik und Informatik > Informatik > Datenanalyse (Roth)
UniBasel Contributors:Roth, Volker and Keller, Sebastian Mathias and Samarin, Maxim and Wieser, Mario
Item Type:Conference or Workshop Item, refereed
Conference or workshop item Subtype:Conference Paper
Publisher:Springer International Publishing
Series Name:Lecture Notes in Computer Science
Issue Number:11824
Note:Publication type according to Uni Basel Research Database: Conference paper
Related URLs:
Identification Number:
Last Modified:10 Mar 2020 16:14
Deposited On:10 Mar 2020 16:14

Repository Staff Only: item control page