AI Paper Atlas

Research map / Architectures and efficiency / Quantization

INSTA-BNN: Binary Neural Network with INSTAnce-aware Threshold

Changhun Lee, Hyungjun Kim, Eunhyeok Park, Jae-Joon Kim

ICCV 2023 · Architectures and efficiency › Quantization · approach: Binary neural networks · 17 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 RSign/RPReLU with INSTA-Th and INSTA-PReLU (ablation)ReActNet-ResNet18 · ImageNet-1k val (from scratch ablation)top-1 accuracy: 60.0 -> 62.1Improved
add instance-wise mean in threshold (ablation)ReActNet-ResNet18 · ImageNet-1k valtop-1 accuracy: 60.0 -> 60.6Improved
add variance and skewness via E[x^3] (ablation)INSTA-Th · ImageNet-1k valtop-1 accuracy: 60.6 -> 61.1Improved
add normalization layer before computing instance statistics (ablation)INSTA-Th · ImageNet-1k valtop-1 accuracy: 60.8 -> 61.1Improved
add SE-like module for alpha (ablation)INSTA-Th · ImageNet-1k valtop-1 accuracy: 61.1 -> 61.7Improved
use SE module alone on baseline instead of INSTA modules (ablation)ReActNet-ResNet18 · ImageNet-1k valtop-1 accuracy: 60.0 -> 60.8Improved
skip INSTA modules in earlier layers (conv2_x) (ablation)INSTA-BNN-ResNet18 · ImageNet-1k valtop-1 accuracy: 62.1 -> 62.1No clear effect
apply INSTA-BNN to ReActNet-A MobileNetV1 backboneReActNet-A · ImageNet-1k valtop-1 accuracy: 69.4 -> 71.7Improved
apply INSTA-BNN to ReActNet-ResNet18 backboneReActNet-ResNet18 · ImageNet-1k valtop-1 accuracy: 65.5 -> 67.6Improved
add INSTA modules on device latencyReActNet-ResNet18 · Pixel 3 A75 latencylatency (ms): 32 -> 34Hurt
use mean, variance and skewness in threshold instead of mean only (ablation)binary ResNet-20 · CIFAR-10top-1 accuracy: 87.32% is about 0.9% higher than ReActNet learnable threshold baselineImproved

Later papers that compared with this method

1 later accepted paper reports INSTA-BNN as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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