AI Paper Atlas

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

Improved Conditional VRNNs for Video Prediction

Lluis Castrejon, Nicolas Ballas, Aaron Courville

ICCV 2019 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Hierarchical VAEs and posterior collapse · 147 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
add ConvLSTM layers to the decoder (1 to 3) (ablation)VRNN likelihood · BAIR Pushtest ELBO: 3826 -> 2355Improved
add ConvLSTM layers to the decoder (3 to 6) (ablation)VRNN likelihood · BAIR Pushtest ELBO: 2355 -> 1731.31Improved
increase ConvLSTM hidden size of the 6-layer decoder (ablation)VRNN likelihood · BAIR Pushtest ELBO: 1731.31 -> 1589.72Improved
add a second latent level at 8x8 (ablation)VRNN with 1 latent level · BAIR Pushtest ELBO: 1536.93 -> 1313.02Improved
add a third latent level at 32x32 (1-8-32) (ablation)VRNN with 2 latent levels · BAIR Pushtest ELBO: 1313.02 -> 1162.24Improved
add a fourth latent level at 16x16 (1-8-16-32) (ablation)1-8-32 VRNN · BAIR Pushtest ELBO: 1162.24 -> 1256.22Hurt
add beta warmup and dense connectivity together (ablation)1-8-32 VRNN · BAIR Pushtest ELBO: 1440.58 -> 1162.24Improved
add beta warmup alone (ablation)1-8-32 VRNN · BAIR Pushtest ELBO: 1440.58 -> 1440.62No clear effect
add dense connectivity alone (ablation)1-8-32 VRNN · BAIR Pushtest ELBO: 1440.58 -> 1547.05Hurt
add latent hierarchy to single-level VRNNOurs · Stochastic Moving MNISTFVD: 63.81 -> 57.17Improved
add latent hierarchy to single-level VRNNOurs · Stochastic Moving MNISTSSIM: 0.763 -> 0.760No clear effect
replace SVG-LP with higher-capacity VRNNSVG-LP · Stochastic Moving MNISTFVD: 90.81 -> 63.81Improved

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