AI Paper Atlas

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

A Comprehensive Overhaul of Feature Distillation

Byeongho Heo, Jeesoo Kim, Sangdoo Yun, Hyojin Park, Nojun Kwak, Jin Young Choi

ICCV 2019 · Label-efficient and robust learning › Knowledge distillation · approach: Knowledge distillation for detectors · 740 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
distill at pre-ReLU position instead of end of block (ablation)feature distillation baseline · CIFAR-100 setup (c)error rate (%): 26.37 -> 24.81Improved
use teacher batch norm in training mode (BN student transform) (ablation)feature distillation baseline · CIFAR-100 setup (c)error rate (%): 24.81 -> 24.68Improved
add margin ReLU teacher transform and partial L2 distance (ablation)feature distillation baseline · CIFAR-100 setup (c)error rate (%): 24.68 -> 24.08Improved
teacher batch norm in training mode rather than evaluation mode (ablation)Proposed · CIFAR-100 setup (c)error rate (%): 24.54 -> 24.08Improved
teacher batch norm in training mode rather than evaluation mode (ablation)FitNets · CIFAR-100 setup (c)error rate (%): 26.92 -> 26.61Improved
teacher batch norm in training mode rather than evaluation mode (ablation)AT · CIFAR-100 setup (c)error rate (%): 26.42 -> 26.56Hurt
teacher batch norm in training mode rather than evaluation mode (ablation)KD · CIFAR-100 setup (c)error rate (%): 26.45 -> 26.47No clear effect
move distillation to pre-ReLU position (ablation)FT · CIFAR-100 setup (c)error rate (%): 25.91 -> 25.11Improved
move distillation to pre-ReLU position (ablation)Jacobian · CIFAR-100 setup (c)error rate (%): 26.71 -> 26.27Improved
move distillation to pre-ReLU position (ablation)FitNets · CIFAR-100 setup (c)error rate (%): 26.30 -> 26.22No clear effect
move distillation to pre-ReLU position (ablation)AT · CIFAR-100 setup (c)error rate (%): 26.42 -> 26.45No clear effect
use multiple distillation layers instead of a single layer (ablation)Proposed · CIFAR-100 setup (e)error rate (%): 18.17 -> 17.80No 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