AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

Self-Regulated Feature Learning via Teacher-Free Feature Distillation

Lujun Li

ECCV 2022 · Label-efficient and robust learning › Knowledge distillation · approach: Rethinking distillation objectives · 66 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
add all intra-layer Tf-FD losses on top of nothing (ablation)ResNet-20 · CIFAR-100top-1 accuracy (%): 69.06 -> 70.51Improved
add intra-layer Tf-FD at the deepest stage S4 rather than S2 (ablation)ResNet-20 · CIFAR-100top-1 accuracy (%): 70.26 -> 70.41Improved
add all three inter-layer losses (ablation)ResNet-20 · CIFAR-100top-1 accuracy (%): 69.06 -> 70.31Improved
add intra-layer loss to inter-layer losses (full Tf-FD) (ablation)ResNet-20 · CIFAR-100top-1 accuracy (%): 70.31 -> 70.62Improved
add a single Tf-FD loss (ablation)ResNet-20 · CIFAR-100top-1 accuracy (%): 0.78 ~ 1.23 gainsImproved
add logits KD to Tf-FD (Tf-FD†)ResNet-20 · CIFAR-100top-1 accuracy (%): 70.62 -> 71.56Improved
replace DropBlock feature regularization with Tf-FDResNet-20 · CIFAR-100top-1 accuracy (%): 69.65 -> 70.62Improved
add Tf-FD to ResNet-18ResNet-18 · ImageNet valtop-1 accuracy (%): 69.75 -> 70.46Improved
add Tf-FD† to ResNet-18ResNet-18 · ImageNet valtop-1 accuracy (%): 69.75 -> 71.00Improved
add Tf-FD† to Faster R-CNN R50-FPNFaster R-CNN R50-FPN · MS-COCO2017 valAP: 37.93 -> 38.92Improved
replace l2 mimicking loss with AT, SP or ICKD (ablation)Tf-FD ResNet-20 · CIFAR-100top-1 accuracy (%)Improved
replace l2 mimicking loss with AB (ablation)Tf-FD ResNet-20 · CIFAR-100top-1 accuracy (%)Hurt

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