AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Bayes-Factor-VAE: Hierarchical Bayesian Deep Auto-Encoder Models for Factor Disentanglement

Minyoung Kim, Yuting Wang, Pritish Sahu, Vladimir Pavlovic

ICCV 2019 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Hierarchical VAEs and posterior collapse · 25 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use BF-VAE-2 hierarchical prior with relevance indicators instead of F-VAEFactor-VAE · SpritesMetric II accuracy: 77.6 -> 85.9Improved
use BF-VAE-2 prior instead of F-VAEFactor-VAE · O-SprMetric II accuracy: 80.6 -> 97.1Improved
use BF-VAE-1 prior instead of F-VAEFactor-VAE · TeapotsMetric II accuracy: 74.6 -> 82.7Improved
use BF-VAE-1 hierarchical Gamma prior instead of F-VAEFactor-VAE · 3D-FaceMetric II accuracy: 92.8 -> 97.2Improved
use BF-VAE-1 instead of RF-VAE (relevance without explicit non-Gaussianity)RF-VAE · SpritesMetric I accuracy: 85.4 -> 93.8Improved
use BF-VAE-2 instead of RF-VAERF-VAE · C-SprMetric I accuracy: 80.7 -> 94.7Improved
use BF-VAE-1 or BF-VAE-2 instead of RF-VAERF-VAE · TeapotsMetric I accuracy: 98.7 vs 97.6 and 97.9No clear effect
use BF-VAE-0 instead of BF-VAE-1BF-VAE · TeapotsMetric II accuracy: 82.7 -> 90.4Improved
replace the Gaussian prior with a high-capacity Gaussian mixture prior in F-VAE (ablation)Factor-VAE · O-Spr and SpritesMetric II accuracyHurt
increase mixture components K of the MoG prior (ablation)Factor-VAE with MoG prior · O-Spr and SpritesMetric II accuracyHurt
increase the trade-off parameter eta (more weight on sparsity of relevance) (ablation)BF-VAE-2 · Celeb-Anumber of detected relevant factorsImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy