AI Paper Atlas

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

AMLN: Adversarial-based Mutual Learning Network for Online Knowledge Distillation

Xiaobing Zhang, Shijian Lu, Haigang Gong, Zhipeng Luo, Ming Liu

ECCV 2020 · Label-efficient and robust learning › Knowledge distillation · approach: Rethinking distillation objectives · 18 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 block-wise intermediate-to-final feature alignment loss L_F on top of CE+KL (ablation)AMLN ResNet32 peers · CIFAR-100top-1 accuracy: 70.98 -> 73.24Improved
add block-wise intermediate-to-final feature alignment loss L_F on top of CE+KL (ablation)AMLN ResNet32 peers · CIFAR-10top-1 accuracy: 93.15 -> 93.89Improved
add adversarial block-wise discriminator loss L_D on top of CE+KL (ablation)AMLN ResNet32 peers · CIFAR-100top-1 accuracy: 70.98 -> 74.16Improved
add adversarial block-wise discriminator loss L_D on top of CE+KL (ablation)AMLN ResNet32 peers · CIFAR-10top-1 accuracy: 93.15 -> 94.01Improved
add L_D and L_F together to outcome-driven losses (ablation)AMLN ResNet32 peers · CIFAR-100top-1 accuracy: 70.98 -> 74.69Improved
add L_D and L_F together to outcome-driven losses (ablation)AMLN ResNet32 peers · CIFAR-10top-1 accuracy: 93.15 -> 94.25Improved
add L_F on top of L_D (full vs D only) (ablation)AMLN ResNet32 peers · CIFAR-100top-1 accuracy: 74.16 -> 74.69Improved
add KL mutual distillation to CE (DML-like) (ablation)ResNet32 peers · CIFAR-10top-1 accuracy: 93.10 -> 93.15No clear effect
add KL mutual distillation to CE (DML-like) (ablation)ResNet32 peers · CIFAR-100top-1 accuracy: 69.71 -> 70.98Improved
replace FFL with AMLN online distillationResNet34 peers · ImageNettop-1 accuracy: 74.42 -> 75.48Improved
replace FFL with AMLN online distillationResNet32 peers · CIFAR-100top-1 accuracy: 72.97 -> 74.69Improved
replace FFL with AMLN online distillationWRN-40-2 peers · CIFAR-10top-1 accuracy: 95.17 -> 95.21Improved

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