Variational Autoencoders (VAEs) are powerful generative models that merge elements from statistics and information theory with the fexibility ofered by deep neural networks to efciently solve the generation problem for high-dimensional data. The key insight of VAEs is to learn the latent distribution of data in such a way that new meaningful samples can be generated from it. This approach led to tremendous research and variations in the architectural design of VAEs, nourishing the recent feld of research known as unsupervised representation learning. In this article, we provide a comparative evaluation of some of the most successful, recent variations of VAEs. We particularly focus the analysis on the energetic efciency of the diferent models, in the spirit of the so-called Green AI, aiming both to reduce the carbon footprint and the fnancial cost of generative techniques. For each architecture, we provide its mathematical formulation, the ideas underlying its design, a detailed model description, a running implementation and quantitative results.

A Survey on Variational Autoencoders from a Green AI Perspective / Asperti, Andrea; Evangelista, Davide; Loli Piccolomini, Elena. - In: SN COMPUTER SCIENCE. - ISSN 2661-8907. - ELETTRONICO. - 2:4(2021), pp. 301.1-301.23. [10.1007/s42979-021-00702-9]

A Survey on Variational Autoencoders from a Green AI Perspective

Asperti, Andrea
;
Evangelista, Davide;Loli Piccolomini, Elena
2021

Abstract

Variational Autoencoders (VAEs) are powerful generative models that merge elements from statistics and information theory with the fexibility ofered by deep neural networks to efciently solve the generation problem for high-dimensional data. The key insight of VAEs is to learn the latent distribution of data in such a way that new meaningful samples can be generated from it. This approach led to tremendous research and variations in the architectural design of VAEs, nourishing the recent feld of research known as unsupervised representation learning. In this article, we provide a comparative evaluation of some of the most successful, recent variations of VAEs. We particularly focus the analysis on the energetic efciency of the diferent models, in the spirit of the so-called Green AI, aiming both to reduce the carbon footprint and the fnancial cost of generative techniques. For each architecture, we provide its mathematical formulation, the ideas underlying its design, a detailed model description, a running implementation and quantitative results.
2021
A Survey on Variational Autoencoders from a Green AI Perspective / Asperti, Andrea; Evangelista, Davide; Loli Piccolomini, Elena. - In: SN COMPUTER SCIENCE. - ISSN 2661-8907. - ELETTRONICO. - 2:4(2021), pp. 301.1-301.23. [10.1007/s42979-021-00702-9]
Asperti, Andrea; Evangelista, Davide; Loli Piccolomini, Elena
File in questo prodotto:
File Dimensione Formato  
Asperti2021_Article_ASurveyOnVariationalAutoencode.pdf

accesso aperto

Tipo: Versione (PDF) editoriale
Licenza: Licenza per Accesso Aperto. Creative Commons Attribuzione (CCBY)
Dimensione 5.6 MB
Formato Adobe PDF
5.6 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/11585/821669
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 21
  • ???jsp.display-item.citation.isi??? ND
social impact