AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Learning Without Forgetting

Zhizhong Li, Derek Hoiem

ECCV 2016 · Architectures and efficiency › Pruning and efficient tuning · approach: Parameter-efficient transfer analysis · 4,525 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 LwF response-preserving loss instead of fine-tuningAlexNet ImageNet->CUB · ImageNet->CUB (val)old-task accuracy: -3.8 vs LwF 54.7Improved
add LwF instead of feature extractionAlexNet ImageNet->CUB · ImageNet->CUB (val)new-task accuracy: feature extraction -5.2 vs LwF 57.7Improved
make more layers task-specific (last hidden layer) (ablation)LwF AlexNet · ImageNet->CUB (val)new-task accuracy: 57.7 -> 56.2No clear effect
add network expansion with frozen original weights (ablation)AlexNet · ImageNet->CUB (val)new-task accuracy: 57.7 -> 54.0Hurt
add LwF to network expansion (ablation)network expansion · ImageNet->CUB (val)new-task accuracy: 54.0 -> 57.0Improved
lower shared-parameter learning rate to 10% in fine-tuning (ablation)fine-tuning AlexNet · ImageNet->CUB (val)old-task accuracy: 52.2 vs LwF 54.7No clear effect
add warm-up step to fine-tuning (ablation)fine-tuning AlexNet · ImageNet->CUB (val)old-task accuracy: 42.5 -> 50.9Improved
add warm-up step to LwF (ablation)LwF AlexNet · ImageNet->CUB (val)old-task accuracy: 53.5 -> 54.7No clear effect
regularize outputs (LwF) instead of L2 on parameters (ablation)fine-tuning · ImageNet->Scenes / Places365->VOCaccuracyImproved
use knowledge distillation loss instead of L1, L2 or cross-entropy (ablation)LwF · old/new task accuracyaccuracyNo clear effect
LwF versus joint training on AlexNetLwF · ImageNet->CUB (val)old-task accuracy: 0.6 vs LwF 54.7 (joint minus LwF)Mixed
replace fine-tuning with LwF in MD-Net online trainingMD-Net · VOT 2015expected average overlap: 0.373 -> 0.383No clear effect

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