The reviewed record of science sign in
Pith

arxiv: 2002.05227 · v3 · pith:JHQSKCGB · submitted 2020-02-12 · cs.LG · stat.ML

Variational Autoencoders with Riemannian Brownian Motion Priors

Dimitris Kalatzis , David Eklund , Georgios Arvanitidis , S{\o}ren Hauberg This is my paper

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:JHQSKCGBrecord.jsonopen to challenge →

classification cs.LG stat.ML
keywords priorlatentriemannianassumptionautoencodersbrowniancapacityeuclidean
0
0 comments X
read the original abstract

Variational Autoencoders (VAEs) represent the given data in a low-dimensional latent space, which is generally assumed to be Euclidean. This assumption naturally leads to the common choice of a standard Gaussian prior over continuous latent variables. Recent work has, however, shown that this prior has a detrimental effect on model capacity, leading to subpar performance. We propose that the Euclidean assumption lies at the heart of this failure mode. To counter this, we assume a Riemannian structure over the latent space, which constitutes a more principled geometric view of the latent codes, and replace the standard Gaussian prior with a Riemannian Brownian motion prior. We propose an efficient inference scheme that does not rely on the unknown normalizing factor of this prior. Finally, we demonstrate that this prior significantly increases model capacity using only one additional scalar parameter.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. An Elastic Shape Variational Autoencoder for Skeleton Pose Trajectories

    cs.CV 2026-05 unverdicted novelty 7.0

    ES-VAE learns generative models of skeleton trajectories on elastic shape manifolds by normalizing away rigid motions and timing variability, outperforming standard VAEs on clinical gait prediction and NTU action recognition.

  2. An Elastic Shape Variational Autoencoder for Skeleton Pose Trajectories

    cs.CV 2026-05 unverdicted novelty 6.0

    ES-VAE applies TSRVF representation on Kendall's shape manifold inside a VAE to generate and classify skeletal trajectories while removing rigid transformations and timing variability, showing gains over standard VAEs...

  3. An Elastic Shape Variational Autoencoder for Skeleton Pose Trajectories

    cs.CV 2026-05 unverdicted novelty 5.0

    ES-VAE is a variational autoencoder for skeleton trajectories that uses the transported square-root velocity field on Kendall's shape manifold to remove rigid motions and timing variations, outperforming standard VAEs...