AI Paper Atlas

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

On the Generative Utility of Cyclic Conditionals

Chang Liu, Haoyue Tang, Tao Qin, Jintao Wang, Tie-Yan Liu

NeurIPS 2021 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Score-based generative modeling analysis · 0 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 compatibility loss when training (ablation)CyGen-PT (SGLD) · Synthetic data, iteration 30000compatibility loss: 1.2\e{8} -> 4.6\e{3}Improved
generate data with SGLD instead of Gibbs sampling (ablation)CyGen-PT · Synthetic datageneration quality (qualitative)Improved
pretrain likelihood model as a VAE (CyGen-PT vs CyGen) (ablation)CyGen · Synthetic datageneration quality / posterior centeringNo clear effect
smaller compatibility loss during trainingCyGen · Synthetic datageneration qualityImproved
replace DAE with CyGen-PTdownstream classifier on latent · MNISTaccuracy (%): 98.0 -> 98.3Improved
replace VAE with CyGen-PTdownstream classifier on latent · MNISTaccuracy (%): 94.5 -> 98.3Improved
replace DAE with CyGen-PTdownstream classifier on latent · SVHNaccuracy (%): 74.5 -> 75.8Improved
replace VAE with CyGen-PTdownstream classifier on latent · SVHNaccuracy (%): 30.8 -> 75.8Improved
CyGen-PT vs GibbsNet (reported, different architecture)downstream classifier on latent · SVHNaccuracy (%): 79.6 -> 75.8Hurt
replace DAE or VAE with CyGen-PTgenerative model · SVHNFID: 157 -> 102Improved
replace VAE with CyGen-PTgenerative model · SVHNFID: 128 -> 102Improved
use BiGAN/GibbsNet with the same probabilistic flow architecturebaseline generative models · MNIST and SVHNgeneration qualityHurt

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