AI Paper Atlas

Research map / Continual learning / Online and replay-based

NPCL: Neural Processes for Uncertainty-Aware Continual Learning

Saurav Jha, Dong Gong, He Zhao, Lina Yao

NeurIPS 2023 · Continual learning › Online and replay-based · approach: Unified and unsupervised continual frameworks · 3 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 hierarchical global and task-specific latent variables with distribution regularization and entropy-based head selectionNPCL · S-CIFAR-10, memory 200accuracy: 61.93 ± 1.79 -> 63.78 ± 1.70Improved
Use hierarchical global and task-specific latent variables with distribution regularization and entropy-based head selectionNPCL · S-CIFAR-100, memory 2000accuracy: 51.89 -> 46.71Hurt
Use hierarchical global and task-specific latent variables with distribution regularization and entropy-based head selectionNPCL · S-Tiny-ImageNet, memory 500accuracy: 17.75 ± 1.14 -> 15.29 ± 1.02Hurt
Use hierarchical global and task-specific latent variables with distribution regularization and entropy-based head selectionNPCL · P-MNIST, memory 200accuracy: 81.74 ± 1.07 -> 83.11 ± 0.90Improved
Use hierarchical global and task-specific latent variables with distribution regularization and entropy-based head selectionNPCL · R-MNIST, memory 500accuracy: 92.24 ± 1.12 -> 92.07 ± 1.39Hurt
Add global and task-specific distribution regularization (ablation)NPCL · S-CIFAR-10accuracy: 32.24 -> 63.78Improved
Add task-specific distribution regularization (ablation)NPCL · S-CIFAR-10accuracy: 32.24 -> 57.28Improved
Add global distribution regularization (ablation)NPCL · S-CIFAR-10accuracy: 32.24 -> 50.68Improved
Remove global latent and inter-task hierarchy (ablation)NPCL · S-CIFAR-10, memory 200accuracy: 63.78 ± 1.70 -> 54.6 ± 2.14Hurt
Average logits across task heads rather than selecting the lowest-entropy head (ablation)NPCL · S-CIFAR-10, memory 200accuracy: 63.78 ± 1.70 -> 19.54 ± 3.44Hurt
Use NPCL rather than DERcontinual learning · S-CIFAR-10, memory 200ECE: 0.2991 -> 0.2103Improved
Use NPCL rather than DERfew-shot replay · S-CIFAR-100, memory 5accuracy: 8.96 -> 10.22Improved

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