Hyperspherical Variational Auto-Encoders

Code

• github.com/nicola-decao/s-vae-pytorch
  OfficialIn paperpytorch★ 388
• github.com/nicola-decao/s-vae-tf
  OfficialIn papertf★ 0
• github.com/nicola-decao/s-vae
  OfficialIn papertf★ 0
• github.com/clementchadebec/benchmark_VAE
  pytorch★ 1,983
• github.com/Chenxingyu1990/gzsl_svae
  pytorch★ 28
• github.com/Chenxingyu1990/A-Boundary-Based-Out-of-Distribution-Classifier-for-Generalized-Zero-Shot-Learning
  pytorch★ 28
• github.com/jiacheng-xu/vmf_vae_nlp
  pytorch★ 0
• github.com/acr42/Neural-Variational-Knowledge-Graphs
  tf★ 0
• github.com/alexanderimanicowenrivers/Neural-Variational-Knowledge-Graphs
  tf★ 0

Abstract

The Variational Auto-Encoder (VAE) is one of the most used unsupervised machine learning models. But although the default choice of a Gaussian distribution for both the prior and posterior represents a mathematically convenient distribution often leading to competitive results, we show that this parameterization fails to model data with a latent hyperspherical structure. To address this issue we propose using a von Mises-Fisher (vMF) distribution instead, leading to a hyperspherical latent space. Through a series of experiments we show how such a hyperspherical VAE, or S-VAE, is more suitable for capturing data with a hyperspherical latent structure, while outperforming a normal, N-VAE, in low dimensions on other data types. Code at http://github.com/nicola-decao/s-vae-tf and https://github.com/nicola-decao/s-vae-pytorch

Tasks

Benchmark Results

Reproductions