Hyperspherical Variational Auto-Encoders

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... (read more)

PDF Abstract

Datasets


Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Link Prediction Citeseer S-VGAE AUC 94.7% # 4
AP 95.2% # 4
Link Prediction Cora S-VGAE AUC 94.1% # 2
AP 94.1% # 2
Link Prediction Pubmed S-VGAE AUC 96.0% # 3
AP 96.0% # 3

Methods used in the Paper


METHOD TYPE
VAE
Generative Models