AI Paper Atlas

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

Variational Walkback: Learning a Transition Operator as a Stochastic Recurrent Net

Anirudh Goyal ALIAS PARTH GOYAL, Nan Rosemary Ke, Surya Ganguli, Yoshua Bengio

NeurIPS 2017 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Posterior sampling with diffusion guidance · 11 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
Increase MNIST walkback training from 5 to 15 steps (ablation)VW · MNISTlog-likelihood lower bound: 510 -> 1018Improved
Increase MNIST walkback training from 5 to 10 steps (ablation)VW · MNISTlog-likelihood lower bound: 510 -> 801Improved
Use importance sampling rather than the lower-bound estimate (ablation)VW · MNISTtest log-likelihood: 1018 -> 1243Improved
Use VW instead of NETgenerative modeling · CIFAR-10 testbits/dim: 5.40 -> 4.40Improved
Use VW instead of VAE-10generative modeling · MNISTtest log-likelihood importance sampling estimate: 705 ± 7.4 -> 712 ± 3.4Improved
Use VW instead of VAE-50generative modeling · MNISTtest log-likelihood importance sampling estimate: 991 ± 6.5 -> 712 ± 3.4Hurt
Use layer normalization in the MNIST bottleneck instead of batch normalization (ablation)VW encoder-decoder · MNISTsample qualityImproved
Add dropout (ablation)VW encoder-decoder · MNISTsample qualityImproved
Increase walkback training trajectory length (ablation)VW · MNISTlog-likelihood lower bound: 702 -> 890 -> 1148Improved

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