Research map / Architectures and efficiency / Quantization
ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions
Zechun Liu, Zhiqiang Shen, Marios Savvides, 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 concatenated duplicated downsampling blocks (ablation) | baseline network · ImageNet val | top-1 accuracy: 59.6 -> 62.5 | Improved |
| add distributional loss (ablation) | proposed baseline network · ImageNet val | top-1 accuracy: 61.1 -> 62.5 | Improved |
| add distributional loss (ablation) | baseline without concatenation · ImageNet val | top-1 accuracy: 58.2 -> 59.6 | Improved |
| add RSign (ablation) | proposed baseline network · ImageNet val | top-1 accuracy: 62.5 -> 66.1 | Improved |
| add RPReLU (ablation) | proposed baseline network · ImageNet val | top-1 accuracy: 62.5 -> 67.4 | Improved |
| add both RSign and RPReLU (ablation) | proposed baseline network · ImageNet val | top-1 accuracy: 62.5 -> 69.4 | Improved |
| use RSign and RPReLU instead of Sign and PReLU (ablation) | proposed baseline network · ImageNet val | top-1 accuracy: 65.5 -> 69.4 | Improved |
| add PReLU (ablation) | proposed baseline network · ImageNet val | top-1 accuracy: 62.5 -> 65.5 | Improved |
| apply ReAct operations (ablation) | Bi-RealNet-18 · ImageNet val | top-1 accuracy: 56.4 -> 65.5 | Improved |
| replace 1-bit downsampling convs with group real-valued convs | ReActNet-A · ImageNet val | top-1 accuracy: 69.4 -> 70.1 | Improved |
| replace 1-bit downsampling convs with full real-valued convs | ReActNet-A · ImageNet val | top-1 accuracy: 69.4 -> 71.4 | Improved |
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.
- Bi-RealNet-18 — Bi-Real Net: Enhancing the Performance of 1-bit CNNs with Improved Representational Capability and Advanced Training Algorithm (ECCV 2018) · better on 4 of 4 shared results
- XNOR-Net — XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks (ECCV 2016) · better on 1 of 1 shared result
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
- Training binary neural networks with real-to-binary convolutions (ICLR 2020 · similarity 0.91)
- Bi-Real Net: Binarizing Deep Network Towards Real-Network Performance (IJCV 2019 · similarity 0.89)
- How to Train a Compact Binary Neural Network with High Accuracy? (AAAI 2017 · similarity 0.89)
- Sparsity-Inducing Binarized Neural Networks (AAAI 2020 · similarity 0.88)
- Bi-Real Net: Enhancing the Performance of 1-bit CNNs with Improved Representational Capability and Advanced Training Algorithm (ECCV 2018 · similarity 0.88)
- Layerwise Progressive Freezing: A Training Scaffold for Depth-Scalable Binary Networks (NeurIPS 2026 · similarity 0.86)
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.
