AI Paper Atlas

Research map / Continual learning / Online and replay-based

Learning Dynamic Latent Spaces for Lifelong Generative Modelling

Fei Ye, Adrian G. Bors

AAAI 2023 · Continual learning › Online and replay-based · approach: Unified and unsupervised continual frameworks · 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
Use recursive latent-space expansion with EGAM attentionORVAE · Split CIFAR10FID: 146.30 -> 122.73Improved
Use recursive latent-space expansion with EGAM attentionORVAE · Split TinyImageNetFID: 170.62 -> 131.46Improved
Use ORVAE for classificationORVAE · Split CIFAR10classification accuracy: 48.92 ± 1.32 -> 50.43 ± 0.15Improved
Use ORVAE for classificationORVAE · Split CIFAR100classification accuracy: 21.62 ± 0.69 -> 22.83 ± 0.25Improved
Use ORVAE for density estimationORVAE · Split MNIST-Fashionsample log-likelihood: -242.54 -> -241.40Improved
Use importance-weighted objectives rather than ELBO (ablation)ORVAE · Split MNISTsample log-likelihood: -108.21 -> -120.12; -108.21 -> -133.45Hurt
Use importance-weighted objectives rather than ELBO (ablation)ORVAE · Split MNIST-Fashionsample log-likelihood: -241.40 -> -241.70; -241.40 -> -246.48Hurt
Use a larger memory buffer (ablation)ORVAE · Split MNISTsample log-likelihoodMixed
Increase the expansion threshold λ (ablation)ORVAE · Split MNISTsample log-likelihoodMixed

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