AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Sub-Bit Neural Networks: Learning To Compress and Accelerate Binary Neural Networks

Yikai Wang, Yi Yang, Fuchun Sun, Anbang Yao

ICCV 2021 · Architectures and efficiency › Quantization · approach: Binary neural networks · 19 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 subsets refinement (ablation)Vanilla-SNN 0.56-bit · ImageNet ResNet-18 (W/A 1/32)top-1 accuracy: 62.8 -> 63.4Improved
add subsets refinement (ablation)Vanilla-SNN 0.44-bit · ImageNet ResNet-34 (W/A 1/1)top-1 accuracy: 58.1 -> 58.6Improved
add subsets refinement (ablation)Vanilla-SNN 0.44-bit · CIFAR10 ResNet-20 (W/A 0.44/1)top-1 accuracy: 82.0 -> 82.5Improved
use layer-specific subsets instead of layer-shared (ablation)SNN 0.11-bit · CIFAR10 ResNet-20 (W/A 0.11/32)top-1 accuracy: 73.5 -> 81.6Improved
use layer-specific subsets instead of layer-shared (ablation)Vanilla-SNN 0.56-bit · CIFAR10 ResNet-20 (W/A 0.56/32)top-1 accuracy: 82.6 -> 87.8Improved
use layer-specific subsets instead of layer-shared (ablation)SNN 0.78-bit · CIFAR10 ResNet-20 (W/A 0.78/32)top-1 accuracy: 90.0 -> 90.2Improved
reduce bit-width from 0.67 to 0.44SNN · ImageNet ResNet-18 (W/A 1/32)top-1 accuracy: 64.7 -> 60.9Hurt
replace 1-bit IR-Net with 0.56-bit SNNIR-Net · ImageNet ResNet-18 (W/A 1/32)top-1 accuracy: 66.5 -> 63.4Hurt
replace 1-bit BNN with 0.56-bit SNN1-bit BNN · FPGA ResNet-18running time: 3.626 ms -> 1.159 ms (3.13x)Improved
replace 1-bit BNN with 0.56-bit SNN1-bit BNN · FPGA ResNet-34running time: 7.753 ms -> 2.329 ms (3.33x)Improved
select kernels uniformly by index (Baseline A) instead of Vanilla-SNN (ablation)Vanilla-SNN · ImageNet ResNet-18 (W/A 0.56/32)top-1 accuracy: 62.8 -> 60.5Hurt
select top frequent kernels each iteration (Baseline B) instead of Vanilla-SNN (ablation)Vanilla-SNN · ImageNet ResNet-18 (W/A 0.56/32)top-1 accuracy: 62.8 -> 62.7No 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