Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External Features

Adrien Bibal; Rebecca Marion; Benoît Frénay

Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External Features

Adrien Bibal, Rebecca Marion, Benoît Frénay

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférence › Article dans les actes d'une conférence/un colloque

12 Téléchargements (Pure)

Résumé

One approach to interpreting multidimensional scaling (MDS)
embeddings is to estimate a linear relationship between the MDS dimen-
sions and a set of external features. However, because MDS only preserves
distances between instances, the MDS embedding is invariant to rotation.
As a result, the weights characterizing this linear relationship are arbitrary
and difficult to interpret. This paper proposes a procedure for selecting
the most pertinent rotation for interpreting a 2D MDS embedding.

langue originale	Anglais
titre	26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Sous-titre	ESANN 2018 : Bruges, Belgium, April 25, 26, 27, 2018
rédacteurs en chef	Michel Verleysen
Lieu de publication	Louvain-la-Neuve
Editeur	CIACO
Pages	537-542
ISBN (Electronique)	9782875870476
Etat de la publication	Publié - 1 janv. 2018
Evénement	26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018) - Bruges, Bruges, Belgique Durée: 25 avr. 2018 → 27 avr. 2018

Une conférence

Une conférence	26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018)
Pays/Territoire	Belgique
La ville	Bruges
période	25/04/18 → 27/04/18

Accès au document

2018-Bibal_Marion-Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External FeaturesAccepted author manuscript, 329 KB

Autres fichiers et liens

Link to publication in Scopus

Interpretability and Explainability in Machine Learning and their Application to Nonlinear Dimensionality Reduction
Auteur: Bibal, A., 16 nov. 2020
Superviseur: FRENAY, B. (Promoteur), VANHOOF, W. (Président), Cleve, A. (Jury), Dumas, B. (Jury), Lee, J. A. (Personne externe) (Jury) & Galarraga, L. A. (Personne externe) (Jury)
Student thesis: Doc types › Docteur en Sciences

Fichier

Contient cette citation

Bibal, Adrien ; Marion, Rebecca ; Frénay, Benoît. / Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External Features. 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning : ESANN 2018 : Bruges, Belgium, April 25, 26, 27, 2018. Editeur / Michel Verleysen. Louvain-la-Neuve : CIACO , 2018. p. 537-542

@inproceedings{c711cd775ac74eeab209e495a3e8a24a,

title = "Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External Features",

abstract = "One approach to interpreting multidimensional scaling (MDS)embeddings is to estimate a linear relationship between the MDS dimen-sions and a set of external features. However, because MDS only preservesdistances between instances, the MDS embedding is invariant to rotation.As a result, the weights characterizing this linear relationship are arbitraryand difficult to interpret. This paper proposes a procedure for selectingthe most pertinent rotation for interpreting a 2D MDS embedding.",

keywords = "Machine learning, Interpretability, Dimensionality reduction, Multidimensional scaling, Multi-view, Sparsity, Lasso regularization",

author = "Adrien Bibal and Rebecca Marion and Beno{\^i}t Fr{\'e}nay",

note = "Funding Information: The authors want to thank Nathan Nguyen from the Universite catholique de Louvain for having pointed to the need for this kind of procedure in psychology. The second author gratefully acknowledges financial support from the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy), the Fonds special de recherche (Federation Wallonie-Bruxelles) and the Belgian Fund for Scientific Research (F.R.S.-FNRS, FRIA grant). Funding Information: The authors want to thank Nathan Nguyen from the Universit{\'e} catholique de Louvain for having pointed to the need for this kind of procedure in psychology. The second author gratefully acknowledges financial support from the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy), the Fonds sp{\'e}cial de recherche (F{\'e}d{\'e}ration Wallonie-Bruxelles) and the Belgian Fund for Scientific Research (F.R.S.-FNRS, FRIA grant). Publisher Copyright: {\textcopyright} ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning.; 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018) ; Conference date: 25-04-2018 Through 27-04-2018",

year = "2018",

month = jan,

day = "1",

language = "English",

pages = "537--542",

editor = "Michel Verleysen",

booktitle = "26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

publisher = "CIACO ",

}

Bibal, A, Marion, R & Frénay, B 2018, Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External Features. Dans M Verleysen (Ed.), 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning : ESANN 2018 : Bruges, Belgium, April 25, 26, 27, 2018. CIACO , Louvain-la-Neuve, p. 537-542, 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018), Bruges, Belgique, 25/04/18.

Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External Features. / Bibal, Adrien; Marion, Rebecca; Frénay, Benoît.

26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning : ESANN 2018 : Bruges, Belgium, April 25, 26, 27, 2018. Ed. / Michel Verleysen. Louvain-la-Neuve : CIACO , 2018. p. 537-542.

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférence › Article dans les actes d'une conférence/un colloque

TY - GEN

T1 - Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External Features

AU - Bibal, Adrien

AU - Marion, Rebecca

AU - Frénay, Benoît

N1 - Funding Information: The authors want to thank Nathan Nguyen from the Universite catholique de Louvain for having pointed to the need for this kind of procedure in psychology. The second author gratefully acknowledges financial support from the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy), the Fonds special de recherche (Federation Wallonie-Bruxelles) and the Belgian Fund for Scientific Research (F.R.S.-FNRS, FRIA grant). Funding Information: The authors want to thank Nathan Nguyen from the Université catholique de Louvain for having pointed to the need for this kind of procedure in psychology. The second author gratefully acknowledges financial support from the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy), the Fonds spécial de recherche (Fédération Wallonie-Bruxelles) and the Belgian Fund for Scientific Research (F.R.S.-FNRS, FRIA grant). Publisher Copyright: © ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - One approach to interpreting multidimensional scaling (MDS)embeddings is to estimate a linear relationship between the MDS dimen-sions and a set of external features. However, because MDS only preservesdistances between instances, the MDS embedding is invariant to rotation.As a result, the weights characterizing this linear relationship are arbitraryand difficult to interpret. This paper proposes a procedure for selectingthe most pertinent rotation for interpreting a 2D MDS embedding.

AB - One approach to interpreting multidimensional scaling (MDS)embeddings is to estimate a linear relationship between the MDS dimen-sions and a set of external features. However, because MDS only preservesdistances between instances, the MDS embedding is invariant to rotation.As a result, the weights characterizing this linear relationship are arbitraryand difficult to interpret. This paper proposes a procedure for selectingthe most pertinent rotation for interpreting a 2D MDS embedding.

KW - Machine learning

KW - Interpretability

KW - Dimensionality reduction

KW - Multidimensional scaling

KW - Multi-view

KW - Sparsity

KW - Lasso regularization

UR - http://www.scopus.com/inward/record.url?scp=85061840537&partnerID=8YFLogxK

M3 - Conference contribution

SP - 537

EP - 542

BT - 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

A2 - Verleysen, Michel

PB - CIACO

CY - Louvain-la-Neuve

T2 - 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018)

Y2 - 25 April 2018 through 27 April 2018

ER -

Finding the Most Interpretable MDS Rotation for Sparse Linear Models based on External Features

Résumé

Une conférence

Accès au document

Autres fichiers et liens

Thèses de l'étudiant

Interpretability and Explainability in Machine Learning and their Application to Nonlinear Dimensionality Reduction

Contient cette citation