Deutsch
 
Hilfe Datenschutzhinweis Impressum
  DetailsucheBrowse

Datensatz

DATENSATZ AKTIONENEXPORT
Zur Ablage hinzufügen
 Lokale TagsFreigabegeschichteDetailsÜbersicht
  Gibbs sampling with people

Harrison, P. M. C., Marjieh, R., Adolfi, F., van Rijn, P., Anglada-Tort, M., Tchernichovski, O., et al. (2021). Gibbs sampling with people. In H. Larochelle, M. Ranzato, R. Hadsell, M. Balcan, & H. Lin (Eds.), 34th Conference on Neural Information Processing Systems (NeurIPS 2020) (pp. 10659-10671). Red Hook, NY: Curran Associates. doi:10.17605/OSF.IO/RZK4S.

Item is

 

einblenden: alle ausblenden: alle

Basisdaten

einblenden: ausblenden:
Datensatz-Permalink: https://hdl.handle.net/21.11116/0000-0007-AB77-3 Versions-Permalink: https://hdl.handle.net/21.11116/0000-000A-DD92-8
Genre: Konferenzbeitrag

Dateien

einblenden: Dateien
ausblenden: Dateien
:
2008.02595.pdf (beliebiger Volltext), 8MB
Öffnen   Speichern
Datei-Permalink:
https://hdl.handle.net/21.11116/0000-0007-AB79-1
Name:
2008.02595.pdf
Beschreibung:
-
OA-Status:
Sichtbarkeit:
Öffentlich
MIME-Typ / Prüfsumme:
application/pdf / [MD5]
Technische Metadaten:
Öffnen
Copyright Datum:
-
Copyright Info:
-
Lizenz:
-

Externe Referenzen

einblenden:

Urheber

einblenden:
ausblenden:
 Urheber:
Harrison, Peter M. C.1, Autor           
Marjieh, Raja1, Autor           
Adolfi, Federico2, Autor           
van Rijn, Pol2, Autor           
Anglada-Tort, Manuel1, Autor           
Tchernichovski, Ofer3, Autor
Larrouy-Maestri, Pauline2, Autor           
Jacoby, Nori1, Autor           
Affiliations:
1Research Group Computational Auditory Perception, Max Planck Institute for Empirical Aesthetics, Max Planck Society, ou_3024247              
2Department of Neuroscience, Max Planck Institute for Empirical Aesthetics, Max Planck Society, ou_2421697              
3Hunter College CUNY, The CUNY Graduate Center, ou_persistent22              

Inhalt

einblenden:
ausblenden:
Schlagwörter: -
 Zusammenfassung: A core problem in cognitive science and machine learning is to understand how humans derive semantic representations from perceptual objects, such as color from an apple, pleasantness from a musical chord, or seriousness from a face. Markov Chain Monte Carlo with People (MCMCP) is a prominent method for studying such representations, in which participants are presented with binary choice trials constructed such that the decisions follow a Markov Chain Monte Carlo acceptance rule. However, while MCMCP has strong asymptotic properties, its binary choice paradigm generates relatively little information per trial, and its local proposal function makes it slow to explore the parameter space and find the modes of the distribution. Here we therefore generalize MCMCP to a continuous-sampling paradigm, where in each iteration the participant uses a slider to continuously manipulate a single stimulus dimension to optimize a given criterion such as 'pleasantness'. We formulate both methods from a utility-theory perspective, and show that the new method can be interpreted as 'Gibbs Sampling with People' (GSP). Further, we introduce an aggregation parameter to the transition step, and show that this parameter can be manipulated to flexibly shift between Gibbs sampling and deterministic optimization. In an initial study, we show GSP clearly outperforming MCMCP; we then show that GSP provides novel and interpretable results in three other domains, namely musical chords, vocal emotions, and faces. We validate these results through large-scale perceptual rating experiments. The final experiments use GSP to navigate the latent space of a state-of-the-art image synthesis network (StyleGAN), a promising approach for applying GSP to high-dimensional perceptual spaces. We conclude by discussing future cognitive applications and ethical implications.

Details

einblenden:
ausblenden:
Sprache(n): eng - English
 Datum: 2020-04-152021
 Publikationsstatus: Erschienen
 Seiten: -
 Ort, Verlag, Ausgabe: -
 Inhaltsverzeichnis: -
 Art der Begutachtung: Interne Begutachtung
 Identifikatoren: DOI: 10.17605/OSF.IO/RZK4S
 Art des Abschluß: -

Veranstaltung

einblenden:
ausblenden:
Titel: Conference on Neural Information Processing Systems ; 34
Veranstaltungsort: Online
Start-/Enddatum: 2020-12-06 - 2020-12-12

Entscheidung

einblenden:

Projektinformation

einblenden:

Quelle 1

einblenden:
ausblenden:
Titel: 34th Conference on Neural Information Processing Systems (NeurIPS 2020)
Genre der Quelle: Konferenzband
 Urheber:
Larochelle , H. , Herausgeber
Ranzato, M., Herausgeber
Hadsell, R., Herausgeber
Balcan, M.F. , Herausgeber
Lin, H. , Herausgeber
Affiliations:
-
Ort, Verlag, Ausgabe: Red Hook, NY : Curran Associates
Seiten: - Band / Heft: 2 Artikelnummer: 894 Start- / Endseite: 10659 - 10671 Identifikator: ISBN: 978-1-7138-2954-6

Quelle 2

einblenden:
ausblenden:
Titel: Advances in neural information processing systems
Genre der Quelle: Reihe
 Urheber:
Affiliations:
Ort, Verlag, Ausgabe: -
Seiten: - Band / Heft: 33 Artikelnummer: - Start- / Endseite: - Identifikator: -