AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Sparsely Aggregated Convolutional Networks

Ligeng Zhu, Ruizhi Deng, Michael Maire, Zhiwei Deng, Greg Mori, Ping Tan

ECCV 2018 · Architectures and efficiency › CNN design · approach: Theory of convolutional inductive bias · 65 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
replace dense aggregation with sparse (logarithmic) aggregation in DenseNet-BC 100 k=32/64/128DenseNet-BC · CIFAR-100+error (%): 18.71 -> 18.22Improved
replace dense aggregation with sparse aggregation in DenseNet-BC (parameters)DenseNet-BC · CIFAR-100+parameters: 30.5M -> 16.7MImproved
replace dense aggregation with sparse aggregation in DenseNet-40 k=12DenseNet · CIFAR-100+error (%): 24.79 -> 24.48Improved
replace dense with sparse aggregation in DenseNet-100 k=12 (similar 7.2M params)DenseNet · CIFAR-100+error (%): 20.97 -> 19.89Improved
replace dense with sparse aggregation in DenseNet-BC-250 k=24DenseNet-BC · CIFAR-100+error (%): 17.6 -> 18.22Hurt
replace dense with sparse aggregation in DenseNet-169-32 on ImageNetDenseNet · ImageNet valtop-1 error (%): 23.6 -> 23.6Mixed
replace dense with sparse aggregation in DenseNet-201-32 on ImageNetDenseNet · ImageNet valtop-1 error (%): 22.6 -> 22.7Mixed
replace dense with sparse aggregation in DenseNet-121-32 on ImageNet (FLOPs)DenseNet · ImageNet valFLOPs: 5.7G -> 3.46GImproved
scale depth from 1001 to 2000 for ResNetResNet · CIFAR-100+error (%): 21.42 -> 22.76Hurt
scale depth from 1001 to 2000 for sparse-sum SparseNetSparseNet[+] · CIFAR-100+error (%): 22.10 -> 21.01Improved
replace summation with concatenation in sparse aggregation (SparseNet[concat]-400 vs SparseNet[+]-200)SparseNet · CIFAR-100+error (%): 25.77 -> 24.53Improved
use sparse instead of dense aggregation at extreme depth (DenseNet-BC k=4 1001 vs SparseNet-BC k=12 400)DenseNet-BC · CIFAR-100+error (%): 28.50 -> 24.53Improved

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