AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

RIFLE: Backpropagation in Depth for Deep Transfer Learning through Re-Initializing the Fully-connected LayEr

Xingjian Li, Haoyi Xiong, Haozhe An, Cheng-Zhong Xu, Dejing Dou

ICML 2020 · Architectures and efficiency › Pruning and efficient tuning · approach: Parameter-efficient transfer analysis · 9 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
Periodically reinitialize the FC layer and use cyclic learning ratesL2 fine-tuning · Image classification transfer benchmarks (8 datasets)top-1 accuracyMixed
Periodically reinitialize the FC layer and use cyclic learning ratesL2-SP fine-tuning · Image classification transfer benchmarks (8 datasets)top-1 accuracyImproved
Remove cyclic learning rate (retain FC reinitialization) (ablation)RIFLE + L2 · Flower-102top-1 accuracy: 93.00$\pm$0.08 -> 92.17$\pm$0.19Hurt
Remove cyclic learning rate (retain FC reinitialization) (ablation)RIFLE + L2 · Food-101-150top-1 accuracy: 76.36$\pm$0.16 -> 76.66$\pm$0.11Improved
Remove FC reinitialization (retain cyclic learning rate) (ablation)RIFLE + L2-SP · Food-101-150top-1 accuracy: 77.36$\pm$0.04 -> 76.44$\pm$0.11Hurt
Remove FC reinitialization (retain cyclic learning rate) (ablation)RIFLE + L2 · Flower-102top-1 accuracy: 93.00$\pm$0.08 -> 92.44$\pm$0.16Hurt

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