AI Paper Atlas

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

ByPE-VAE: Bayesian Pseudocoresets Exemplar VAE

Qingzhong Ai, LIRONG HE, SHIYU LIU, Zenglin Xu

NeurIPS 2021 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Hierarchical VAEs and posterior collapse · 1 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 Gaussian prior with ByPE pseudocoreset priorVAE · Dynamic MNISTtest NLL: 24.41 ± 0.06 -> 23.61 ± 0.03Improved
replace Gaussian prior with ByPE pseudocoreset priorVAE · CIFAR10test NLL: 72.21 ± 0.08 -> 71.91 ± 0.02Improved
ByPE versus VampPriorHVAE · CIFAR10test NLL: 71.54 ± 0.04 -> 71.38 ± 0.01Improved
ByPE versus VampPriorConvHVAE · Dynamic MNISTtest NLL: 22.94 ± 0.02 -> 22.84 ± 0.02Improved
ByPE versus Exemplar priorfully convolutional VAE · CelebAtest NLL: 185.03 ± 1.46 -> 182.11 ± 1.10Improved
ByPE pseudocoreset of 500 instead of 25000 exemplarsExemplar VAE · Dynamic MNISTtraining time (s/epoch): 35.45 -> 13.19Improved
ByPE pseudocoreset of 500 instead of 25000 exemplarsExemplar VAE · CIFAR10training time (s/epoch): 66.85 -> 17.30Improved
ByPE versus Exemplar VAE with 25000 exemplarsByPE-VAE · Fashion MNISTtest NLL: 20.81 vs 20.85No clear effect
update pseudocoreset every k epochs, k from 1 to 100 (ablation)ByPE-VAE · MNISTtest NLL: 23.60 -> 23.70No clear effect
vary lambda mixing weight of real and augmented data from 0 to 0.5 (ablation)ByPE-VAE · permutation invariant MNISTtest error (%): 1.42 ± 0.08 -> 0.83 ± 0.05Improved
raise lambda from 0.5 to 1.0 (ablation)ByPE-VAE · permutation invariant MNISTtest error (%): 0.83 ± 0.05 -> 1.31 ± 0.00Hurt
sample latents from prior instead of posterior for augmentation (ablation)ByPE-VAE · permutation invariant CIFAR10test error (%): 49.00 ± 0.04 -> 46.60 ± 0.13Improved

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