In this paper, we propose the use of data symmetries, in the sense of equivalences under signal transformations, as priors for learning symmetry-adapted data representations, i.e., representations that are equivariant to these transformations. We rely on a group-theoretic definition of equivariance and provide conditions for enforcing a learned representation, for example the weights in a neural network layer or the atoms in a dictionary, to have the structure of a group and specifically the group structure in the distribution of the input. By reducing the analysis of generic group symmetries to permutation symmetries, we devise a regularization scheme for representation learning algorithm, using an unlabeled training set. The proposed regularization is aimed to be a conceptual, theoretical and computational proof of concept for symmetry-adapted representation learning, where the learned data representations are equivariant or invariant to transformations, without explicit knowledge of the underlying symmetries in the data. (C) 2018 Elsevier Ltd. All rights reserved.

Symmetry-adapted representation learning

Fabio Anselmi
;
2019-01-01

Abstract

In this paper, we propose the use of data symmetries, in the sense of equivalences under signal transformations, as priors for learning symmetry-adapted data representations, i.e., representations that are equivariant to these transformations. We rely on a group-theoretic definition of equivariance and provide conditions for enforcing a learned representation, for example the weights in a neural network layer or the atoms in a dictionary, to have the structure of a group and specifically the group structure in the distribution of the input. By reducing the analysis of generic group symmetries to permutation symmetries, we devise a regularization scheme for representation learning algorithm, using an unlabeled training set. The proposed regularization is aimed to be a conceptual, theoretical and computational proof of concept for symmetry-adapted representation learning, where the learned data representations are equivariant or invariant to transformations, without explicit knowledge of the underlying symmetries in the data. (C) 2018 Elsevier Ltd. All rights reserved.
File in questo prodotto:
File Dimensione Formato  
1-s2.0-S0031320318302620-main.pdf

Accesso chiuso

Tipologia: Documento in Versione Editoriale
Licenza: Copyright Editore
Dimensione 1.18 MB
Formato Adobe PDF
1.18 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
1-s2.0-S0031320318302620-main-Post_print.pdf

Open Access dal 10/08/2020

Tipologia: Bozza finale post-referaggio (post-print)
Licenza: Creative commons
Dimensione 1.56 MB
Formato Adobe PDF
1.56 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11368/3035085
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 25
  • ???jsp.display-item.citation.isi??? 16
social impact