We provide a general mathematical framework for group and set equivariance in machine learning. We define group equivariant non-expansive operators (GENEOs) as maps between function spaces associated with groups of transformations. We study the topological and metric properties of the space of GENEOs to evaluate their approximating power and set the basis for general strategies to initialize and compose operators. We define suitable pseudo-metrics for the function spaces, the equivariance groups and the set of non-expansive operators. We prove that, under suitable assumptions, the space of GENEOs is compact and convex. These results provide fundamental guarantees in a machine learning perspective. By considering isometry-equivariant non-expansive operators, we describe a simple strategy to select and sample operators. Thereafter, we show how selected and sampled operators can be used both to perform classical metric learning and to inject knowledge in artificial neural networks.

Bergomi, M.G., Frosini, P., Giorgi, D., Quercioli, N. (2019). Towards a topological–geometrical theory of group equivariant non-expansive operators for data analysis and machine learning. NATURE MACHINE INTELLIGENCE, 1(9), 423-433 [10.1038/s42256-019-0087-3].

Towards a topological–geometrical theory of group equivariant non-expansive operators for data analysis and machine learning

Frosini, Patrizio
Supervision
;
Quercioli, Nicola
2019

Abstract

We provide a general mathematical framework for group and set equivariance in machine learning. We define group equivariant non-expansive operators (GENEOs) as maps between function spaces associated with groups of transformations. We study the topological and metric properties of the space of GENEOs to evaluate their approximating power and set the basis for general strategies to initialize and compose operators. We define suitable pseudo-metrics for the function spaces, the equivariance groups and the set of non-expansive operators. We prove that, under suitable assumptions, the space of GENEOs is compact and convex. These results provide fundamental guarantees in a machine learning perspective. By considering isometry-equivariant non-expansive operators, we describe a simple strategy to select and sample operators. Thereafter, we show how selected and sampled operators can be used both to perform classical metric learning and to inject knowledge in artificial neural networks.
2019
Bergomi, M.G., Frosini, P., Giorgi, D., Quercioli, N. (2019). Towards a topological–geometrical theory of group equivariant non-expansive operators for data analysis and machine learning. NATURE MACHINE INTELLIGENCE, 1(9), 423-433 [10.1038/s42256-019-0087-3].
Bergomi, Mattia G.; Frosini, Patrizio; Giorgi, Daniela; Quercioli, Nicola
File in questo prodotto:
File Dimensione Formato  
Towards_a_topological_geometrical_theory_of_group_equivariant_non_expansive_operators_POSTPRINT.pdf

accesso aperto

Tipo: Postprint
Licenza: Licenza per accesso libero gratuito
Dimensione 3.7 MB
Formato Adobe PDF
3.7 MB Adobe PDF Visualizza/Apri

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/11585/702691
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? 24
social impact