In this thesis, we focused on recent developments in variational inference. We saw how these methods can be married with ideas from deep learning to give life to the example of variational autoencoders (VAEs), models that learn how to generate new realistic-looking data. Along the way, we provided parallels with other statistical methods.

In this thesis, we focused on recent developments in variational inference. We saw how these methods can be married with ideas from deep learning to give life to the example of variational autoencoders (VAEs), models that learn how to generate new realistic-looking data. Along the way, we provided parallels with other statistical methods.

Auto-Encoding Variational Bayes

BRUNO, MATTIA
2021/2022

Abstract

In this thesis, we focused on recent developments in variational inference. We saw how these methods can be married with ideas from deep learning to give life to the example of variational autoencoders (VAEs), models that learn how to generate new realistic-looking data. Along the way, we provided parallels with other statistical methods.
2021
Auto-Encoding Variational Bayes
In this thesis, we focused on recent developments in variational inference. We saw how these methods can be married with ideas from deep learning to give life to the example of variational autoencoders (VAEs), models that learn how to generate new realistic-looking data. Along the way, we provided parallels with other statistical methods.
Variational
Bayesian
Autoencoders
Generative models
Machine learning
File in questo prodotto:
File Dimensione Formato  
Bruno_Mattia.pdf

accesso aperto

Dimensione 3.21 MB
Formato Adobe PDF
3.21 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/38806