Timezone: »

 
Poster
Variational Laplace Autoencoders
Yookoon Park · Chris Kim · Gunhee Kim

Thu Jun 13 06:30 PM -- 09:00 PM (PDT) @ Pacific Ballroom #2
in Posters Thu »

Variational autoencoders employ an amortized inference model to approximate the posterior of latent variables. However, such amortized variational inference faces two challenges: (1) the limited posterior expressiveness of fully-factorized Gaussian assumption and (2) the amortization error of the inference model. We present a novel approach that addresses both challenges. First, we focus on ReLU networks with Gaussian output and illustrate their connection to probabilistic PCA. Building on this observation, we derive an iterative algorithm that finds the mode of the posterior and apply fullcovariance Gaussian posterior approximation centered on the mode. Subsequently, we present a general framework named Variational Laplace Autoencoders (VLAEs) for training deep generative models. Based on the Laplace approximation of the latent variable posterior, VLAEs enhance the expressiveness of the posterior while reducing the amortization error. Empirical results on MNIST, Omniglot, Fashion-MNIST, SVHN and CIFAR10 show that the proposed approach significantly outperforms other recent amortized or iterative methods on the ReLU networks.

Keywords: Approximate Inference · Deep Generative Models

Author Information

Yookoon Park (Seoul National University)
Chris Kim (Seoul National University)
Gunhee Kim (Seoul National University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors