Research map / Optimization / Optimizers
Estimator Meets Equilibrium Perspective: A Rectified Straight Through Estimator for Binary Neural Networks Training
Xiao-Ming Wu, Dian Zheng, Zuhao Liu, Wei-Shi Zheng
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 |
|---|---|---|---|
| replace identity STE with power-function ReSTE estimator (ablation) | binary ResNet-20 · CIFAR-10 1W/1A | Acc (%): 84.44 -> 86.75 | Improved |
| replace RBNN polynomial estimator with ReSTE (ablation) | binary ResNet-20 · CIFAR-10 1W/1A | Acc (%): 85.87 -> 86.75 | Improved |
| increase o_end so estimator approaches sign function | ReSTE · CIFAR-10 1W/1A | estimating error vs gradient instability | Mixed |
| set o_end=10 instead of 3 (ablation) | ReSTE ResNet-20 · CIFAR-10 1W/1A | Acc (%): 86.75 -> 82.86 | Hurt |
| use ReSTE without auxiliary modules or losses instead of RBNN with rotation module | ResNet-20 Bi-Real · CIFAR-10 1W/1A | Acc (%): 87.50 -> 87.92 | Improved |
| use ReSTE instead of FDA with noise adaptation module | ResNet-18 · ImageNet 1W/1A | Top-1 (%): 60.20 -> 60.88 | Improved |
Closest papers by meaning
- AdaSTE: An Adaptive Straight-Through Estimator To Train Binary Neural Networks (CVPR 2022 · similarity 0.88)
- An Empirical study of Binary Neural Networks' Optimisation (ICLR 2019 · similarity 0.88)
- Training Binary Neural Networks via Gaussian Variational Inference and Low-Rank Semidefinite Programming (NeurIPS 2024 · similarity 0.88)
- Understanding Straight-Through Estimator in Training Activation Quantized Neural Nets (ICLR 2019 · similarity 0.87)
- Forward and Backward Information Retention for Accurate Binary Neural Networks (CVPR 2020 · similarity 0.87)
- Layerwise Progressive Freezing: A Training Scaffold for Depth-Scalable Binary Networks (NeurIPS 2026 · similarity 0.85)
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.
