AI Paper Atlas

Research map / Continual learning / Online and replay-based

Continual Variational Autoencoder Learning via Online Cooperative Memorization

Fei Ye, Adrian G. Bors

ECCV 2022 · Continual learning › Online and replay-based · approach: Memory-based and Bayesian online learning · 20 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 random memory removal with OCM selectionsingle VAE · Split MNIST (log-likelihood)log-likelihood: -150.79 -> -132.07Improved
replace random memory with OCMsingle VAE · Split MNIST-Fashion (log-likelihood)log-likelihood: -247.46 -> -215.62Improved
add dynamic expansion to OCMVAE-ELBO-OCM · Cross-domain (log-likelihood)log-likelihood: -201.31 -> -177.29Improved
add IWVAE bound (50 samples) to OCM on a single datasetVAE-ELBO-OCM · Split MNIST (log-likelihood)log-likelihood: -132.07 -> -127.11Improved
add IWVAE bound (50 samples) to OCM on mixed datasetsVAE-ELBO-OCM · Split MNIST-Fashion (log-likelihood)log-likelihood: -215.62 -> -224.34Hurt
add IWVAE bound (50 samples) to OCM on cross-domainVAE-ELBO-OCM · Cross-domain (log-likelihood)log-likelihood: -201.31 -> -204.35Hurt
add OCM to single VAEVAE-ELBO-Random · Split CIFAR10FID: 116.26 -> 98.76Improved
add expansion to OCMVAE-ELBO-OCM · Split CIFAR10FID: 98.76 -> 92.99Improved
add OCM to single VAE on large-scale dataVAE-ELBO-Random · ImageNet 64x64FID: 145.36 -> 133.23Improved
use Dynamic-OCM instead of CoPECoPE · Split CIFAR10 classificationaccuracy: 48.92 -> 49.16Improved
use Dynamic-OCM instead of CNDPMCNDPM · Split CIFAR100 classificationaccuracy: 20.10 -> 21.79Improved
single OCM model instead of expansion models LIMix/CNDPMLIMix/CNDPM · Cross-domain (log-likelihood)log-likelihood: -218.15 -> -201.31Improved

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