AI Paper Atlas

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

Learning Hierarchical Features with Joint Latent Space Energy-Based Prior

Jiali Cui, Ying Nian Wu, Tian Han

ICCV 2023 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Disentangled and discrete latent models · 5 citations (OpenAlex)

Read the paper

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 joint latent EBM prior to hierarchical generator (vs Gaussian-prior VLAE)multi-layer generator · SVHN latent classifiertop-layer accuracy: 52.14% -> 86.63%Improved
increase prior Langevin steps from 15 to 60 (ablation)joint EBM prior · CelebA-64FID: 56.42 -> 32.15Improved
increase Langevin steps beyond 60 (ablation)joint EBM prior · CelebA-64FID: 32.15 -> 30.78No clear effect
increase hidden units of the EBM (nef) from 0 to 100 (ablation)joint EBM prior · SVHNFID: 43.95 -> 24.16Improved
use joint EBM prior instead of baselines for generationgenerator · SVHNFID: 26.19 -> 24.16 vs Multi-NCPImproved
joint EBM prior versus Multi-NCP on SVHN reconstructiongenerator · SVHNMSE: 0.004 (Multi-NCP) vs 0.008 (ours)Hurt
use joint EBM prior for anomaly detection instead of HVAEinference model · MNIST anomaly detectionAUPRC: 0.494 -> 0.722 (digit 1)Improved
use joint EBM prior instead of HVAE with specialized LLR decisionOOD detection · Fashion-MNIST vs MNISTAUPRC: 0.984 (HVAE LLR) vs 0.893 (ours)Hurt
use joint EBM prior versus pro-VLAE with progressive learningdisentanglement · 3DShapesMIG: 0.357 -> 0.554Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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