AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Momentum Auxiliary Network for Supervised Local Learning

Junhao Su, Changpeng Cai, Feiyu Zhu, Chenghao He, Xiaojie Xu, Dongzhi Guan, Chenyang Si

ECCV 2024 · Architectures and efficiency › RNNs and normalization · approach: Differentiable optimization layers · 8 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 EMA of next-block parameters to the auxiliary network (ablation)DGL, ResNet-32 K=16 · CIFAR-10test error (%): 14.08 -> 11.07Improved
add learnable bias on top of EMA (ablation)MAN · CIFAR-10test error (%): 11.07 -> 9.11Improved
add EMA on top of learnable bias for ImageNet (ablation)InfoPro ResNet-101 K=4 · ImageNetTop-1 error (%): 22.81 -> 21.73Improved
use next-block parameters directly instead of EMA (ablation)MAN · CIFAR-10test error (%): 14.08 -> 12.93Improved
use EMA of next-block parameters rather than raw parameters (ablation)MAN · CIFAR-10test error (%): 12.93 -> 11.07Improved
use next-block parameters directly on ImageNet (ablation)InfoPro ResNet-101 K=4 · ImageNetTop-1 error (%): 22.81 -> 22.64No clear effect
add MAN to InfoPro with K=4 blocksInfoPro ResNet-101 · ImageNetTop-1 error (%): 22.81 -> 21.73Improved
add MAN to DGLDGL VGG13 K=10 · ImageNetTop-1 error (%): 35.60 -> 31.99Improved
add MAN to local learning methods on CIFAR-10PredSim, DGL, InfoPro · CIFAR-10 ResNet-32 K=16test error (%): 14.08 -> 9.11Mixed
add MAN to InfoPro for memoryInfoPro ResNet-101 K=4 · ImageNetGPU memory (GB): 10.37 -> 10.69No 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