AI Paper Atlas

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

Copulas as High-Dimensional Generative Models: Vine Copula Autoencoders

Natasa Tagasovska, Damien Ackerer, Thibault Vatter

NeurIPS 2019 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Hierarchical VAEs and posterior collapse · 22 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
Replace the Gaussian copula with a nonparametric copula (ablation)Copula generative model · RingMMD: 0.33(0.02) -> 0.18(0.02)Improved
Replace the Gaussian copula with a nonparametric copula (ablation)Copula generative model · GridMMD: 0.14(0.02) -> 0.15(0.16)Hurt
Replace the Gaussian copula with a nonparametric copula (ablation)Copula generative model · Swiss rollMMD: 0.38(0.02) -> 0.32(0.03)Improved
Replace the Gaussian copula with a nonparametric copula (ablation)Copula generative model · RingNLL: -2.98(0.05) -> -2.47(0.15)Improved
Replace the Gaussian copula with a nonparametric copula (ablation)Copula generative model · Swiss rollCoverage: 0.93(0.03) -> 0.99(0.01)Improved
Use a deeper vine with more trees (ablation)VCAE · FashionMNISTsample qualityImproved
Use a deeper vine with more trees (ablation)VCAE · CelebAsample qualityImproved
Add the clustering-related penalty (ablation)VCAE · SVHNsample qualityNo clear effect
Fit separate vines per class label (ablation)VCAE · MNISTconditional sample qualityMixed
Use VCAE instead of VAEGenerative model · SVHNFID: 0.205 -> 0.167Improved
Use VCAE instead of DCGANGenerative model · SVHNFID: 0.194 -> 0.167Improved
Use VCAE instead of VAEGenerative model · CelebAFID: 0.247 -> 0.233Improved

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