atlas

Research map / Continual learning / Online and replay-based

Uncertainty-based Continual Learning with Adaptive Regularization

NeurIPS 2019 · Continual learning › Online and replay-based · approach: Unified and unsupervised continual frameworks · 113 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Remove upper freeze (ablation)UCL · Split CIFAR-10/100early-task retentionHurt
Remove upper freeze (ablation)UCL · Split CIFAR-10/100average accuracyHelped
Remove component (5) (ablation)UCL · Split CIFAR-10/100Task 1 retentionHurt
Remove component (5) (ablation)UCL · Split CIFAR-10/100average accuracyNo clear effect
Remove component (6) (ablation)UCL · Split CIFAR-10/100Task 1 accuracyHelped
Remove component (6) (ablation)UCL · Split CIFAR-10/100accuracy after Task 1Hurt
Initialize with small uncertainty (σ_init^(l) = 1×10^-3) (ablation)UCL · Eight-task continual reinforcement learningcumulative normalized rewardHurt
Initialize with small uncertainty (σ_init^(l) = 1×10^-3) (ablation)UCL · Eight-task continual reinforcement learning, Task 8new-task learningHurt

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