AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Bi-Real Net: Enhancing the Performance of 1-bit CNNs with Improved Representational Capability and Advanced Training Algorithm

Zechun Liu, Baoyuan Wu, Wenhan Luo, Xin Yang, Wei Liu, Kwang-Ting Cheng

ECCV 2018 · Architectures and efficiency › CNN design · approach: Compact CNN architectures · 589 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 the proposed approximation of the sign derivative (activation backward) (ablation)Bi-Real-18 (ReLU init, original weight update) · ImageNet valtop-1 accuracy (%): 32.9 -> 36.8Improved
add the magnitude-aware weight gradient (weight update) (ablation)Bi-Real-18 (ReLU init) · ImageNet valtop-1 accuracy (%): 32.9 -> 40.5Improved
replace ReLU with Clip for pre-training initialization (ablation)Bi-Real-18 · ImageNet valtop-1 accuracy (%): 32.9 -> 37.4Improved
add all three proposed training techniques (ablation)Bi-Real-18 · ImageNet valtop-1 accuracy (%): 32.9 -> 56.4Improved
add all three proposed training techniques (ablation)Bi-Real-34 · ImageNet valtop-1 accuracy (%): 53.1 -> 62.2Improved
use Bi-Real shortcut structure instead of standard ResNet-18 structure (ablation)binarized ResNet-18 with proposed training · ImageNet valtop-1 accuracy (%): 45.7 -> 56.4Improved
use Bi-Real shortcut structure instead of plain network (ablation)binarized Plain-18 with proposed training · ImageNet valtop-1 accuracy (%): 12.1 -> 56.4Improved
use Bi-Real net instead of XNOR-Net18-layer 1-bit CNN · ImageNet valtop-1 accuracy (%): 51.2% -> 56.4%Improved
use Bi-Real net instead of BinaryNet18-layer 1-bit CNN · ImageNet valtop-1 accuracy (%): 42.2% -> 56.4%Improved
use Bi-Real net instead of ABC-Net34-layer 1-bit CNN · ImageNet valtop-1 accuracy (%): 52.4% -> 62.2%Improved
use Bi-Real net instead of XNOR-Net18-layer 1-bit CNN · ImageNet val costmemory (Mbit): 33.7 Mbit -> 33.6 MbitNo clear effect

Later papers that compared with this method

13 later accepted papers report Bi-Real Net as a baseline in their results tables; 13 of them report a better number on most of the results they share with it.

5 more not listed.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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