AI Paper Atlas

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

Knowledge Transfer via Distillation of Activation Boundaries Formed by Hidden Neurons

Byeongho Heo, Minsik Lee, Sangdoo Yun, Jin Young Choi

AAAI 2019 · Label-efficient and robust learning › Knowledge distillation · approach: Rethinking distillation objectives · 565 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
transfer hidden-neuron activation boundaries with the proposed loss instead of conventional response-transfer methodsWRN16-2 with WRN22-4 teacher · CIFAR-10 test, 1+1 epochserror rate: 43.37% -> 16.39%Improved
transfer hidden-neuron activation boundaries with the proposed lossWRN16-2 with WRN22-4 teacher · CIFAR-10, training data 0.1%error rate: 67.54% -> 50.32%Improved
transfer activation boundaries with the proposed loss rather than the strongest listed baselinestudent networks compressed from larger WRNs · CIFAR-10 tiny networkerror rate: 28.70% -> 23.27%Improved
use activation-boundary transfer with DTL instead of FITNET + DTLMobileNet student with ResNet50 teacher · MIT scenes, 5 examples per classaccuracy: 29.55% -> 43.36%Improved
use activation-boundary transfer with DTL instead of FITNET + DTLMobileNet student with ResNet50 teacher · CUB 2011, 1 example per classaccuracy: 10.70% -> 13.38%Improved
transfer activation boundaries rather than neuron response magnitudes using l2 loss (ablation)WRN16-4 teacher and student · CIFAR-10 classificationerror rate: 10.1% -> 5.1%Improved
transfer activation boundaries rather than neuron response magnitudes using l2 loss (ablation)WRN16-4 teacher and student · CIFAR-10, layer 1 activation similaritysame activation: 56.1% -> 96.3%Improved
set margin μ to 1 instead of 0.75 (ablation)Proposed activation-boundary transfer · CIFAR-10, 10% training dataerror rate: 13.9% -> 13.1%Improved
set margin μ to 1.5 instead of 1 (ablation)Proposed activation-boundary transfer · CIFAR-10, 10% training dataerror rate: 13.1% -> 12.1%Improved
increase margin μ from 1.5 to 2 (ablation)Proposed activation-boundary transfer · CIFAR-10, 10% training dataerror rate: 12.1% -> 12.1%No clear effect
increase margin μ from 2 to 4 (ablation)Proposed activation-boundary transfer · CIFAR-10, 10% training dataerror rate: 12.1% -> 11.7%Improved
transfer responses from multiple layers rather than layer 3 alone (ablation)Proposed + KD · CIFAR-10, 10% training dataerror rate: 17.23% -> 13.09%Improved

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