Fully Bayesian Autoencoders with Latent Sparse Gaussian Processes - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Fully Bayesian Autoencoders with Latent Sparse Gaussian Processes

Résumé

We present a fully Bayesian autoencoder model that treats both local latent variables and global decoder parameters in a Bayesian fashion. This approach allows for flexible priors and posterior approximations while keeping the inference costs low. To achieve this, we introduce an amortized MCMC approach by utilizing an implicit stochastic network to learn sampling from the posterior over local latent variables. Furthermore, we extend the model by incorporating a Sparse Gaussian Process prior over the latent space, allowing for a fully Bayesian treatment of inducing points and kernel hyperparameters and leading to improved scalability. Additionally, we enable Deep Gaussian Process priors on the latent space and the handling of missing data. We evaluate our model on a range of experiments focusing on dynamic representation learning and generative modeling, demonstrating the strong performance of our approach in comparison to existing methods that combine Gaussian Processes and autoencoders.
Fichier principal
Vignette du fichier
tran23a.pdf (1.46 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04188337 , version 1 (25-08-2023)

Identifiants

  • HAL Id : hal-04188337 , version 1

Citer

Ba-Hien Tran, Babak Shahbaba, Stephan Mandt, Maurizio Filippone. Fully Bayesian Autoencoders with Latent Sparse Gaussian Processes. ICML 2023, 40th International Conference on Machine Learning, IEEE, Jul 2023, Honolulu, United States. ⟨hal-04188337⟩
23 Consultations
8 Téléchargements

Partager

Gmail Facebook X LinkedIn More