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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add the proposed approximation of the sign derivative (activation backward) (ablation) | Bi-Real-18 (ReLU init, original weight update) · ImageNet val | top-1 accuracy (%): 32.9 -> 36.8 | Improved |
| add the magnitude-aware weight gradient (weight update) (ablation) | Bi-Real-18 (ReLU init) · ImageNet val | top-1 accuracy (%): 32.9 -> 40.5 | Improved |
| replace ReLU with Clip for pre-training initialization (ablation) | Bi-Real-18 · ImageNet val | top-1 accuracy (%): 32.9 -> 37.4 | Improved |
| add all three proposed training techniques (ablation) | Bi-Real-18 · ImageNet val | top-1 accuracy (%): 32.9 -> 56.4 | Improved |
| add all three proposed training techniques (ablation) | Bi-Real-34 · ImageNet val | top-1 accuracy (%): 53.1 -> 62.2 | Improved |
| use Bi-Real shortcut structure instead of standard ResNet-18 structure (ablation) | binarized ResNet-18 with proposed training · ImageNet val | top-1 accuracy (%): 45.7 -> 56.4 | Improved |
| use Bi-Real shortcut structure instead of plain network (ablation) | binarized Plain-18 with proposed training · ImageNet val | top-1 accuracy (%): 12.1 -> 56.4 | Improved |
| use Bi-Real net instead of XNOR-Net | 18-layer 1-bit CNN · ImageNet val | top-1 accuracy (%): 51.2% -> 56.4% | Improved |
| use Bi-Real net instead of BinaryNet | 18-layer 1-bit CNN · ImageNet val | top-1 accuracy (%): 42.2% -> 56.4% | Improved |
| use Bi-Real net instead of ABC-Net | 34-layer 1-bit CNN · ImageNet val | top-1 accuracy (%): 52.4% -> 62.2% | Improved |
| use Bi-Real net instead of XNOR-Net | 18-layer 1-bit CNN · ImageNet val cost | memory (Mbit): 33.7 Mbit -> 33.6 Mbit | No 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.
- Bi-ViT: Pushing the Limit of Vision Transformer Quantization (AAAI 2024 · better on 5 of 5 shared results)
- PikeLPN: Mitigating Overlooked Inefficiencies of Low-Precision Neural Networks (CVPR 2024 · better on 4 of 4 shared results)
- SA-BNN: State-Aware Binary Neural Network (AAAI 2021 · better on 2 of 2 shared results)
- Effective Training of Convolutional Neural Networks With Low-Bitwidth Weights and Activations (TPAMI 2021 · better on 1 of 1 shared result)
- ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions (ECCV 2020 · better on 4 of 4 shared results)
- Training Quantized Neural Networks With a Full-Precision Auxiliary Module (CVPR 2020 · better on 1 of 1 shared result)
- Training binary neural networks with real-to-binary convolutions (ICLR 2020 · better on 1 of 1 shared result)
- BinaryDuo: Reducing Gradient Mismatch in Binary Activation Network by Coupling Binary Activations (ICLR 2020 · better on 1 of 1 shared result)
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.
- XNOR-Net — XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks (ECCV 2016) · better on 3 of 3 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
- Bi-Real Net: Binarizing Deep Network Towards Real-Network Performance (IJCV 2019 · similarity 0.97)
- Training binary neural networks with real-to-binary convolutions (ICLR 2020 · similarity 0.91)
- BD-Net: Has Depth-Wise Convolution Ever Been Applied in Binary Neural Networks? (AAAI 2026 · similarity 0.89)
- Towards Accurate Binary Convolutional Neural Network (NeurIPS 2017 · similarity 0.88)
- ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions (ECCV 2020 · similarity 0.88)
- Sub-Bit Neural Networks: Learning To Compress and Accelerate Binary Neural Networks (ICCV 2021 · similarity 0.87)
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.
