HEQ自适应调整量化阈值，提升逻辑门残差网络精度与效率
Histogram-Equalized Quantization for logic-gated Residual Neural Networks
- 基于直方图均衡化自动优化量化步长
- CIFAR-10上达顶尖性能，STL-10上精度更高且硬件开销更低
- 适合追求高精度低复杂度部署的模型压缩研究者
在量化神经网络中，根据数据或模型损失调整量化是保证高精度的必要手段。本文提出直方图均衡化量化（HEQ），一种用于线性对称量化的自适应框架。HEQ通过独特的步长优化机制自动调整量化阈值。实验表明，HEQ在CIFAR-10上达到当前最优性能；在STL-10数据集上，更实现了所提逻辑门（OR、MUX）残差网络的稳定训练，精度高于以往工作，且硬件复杂度更低。
原文摘要 · Abstract (English)
Adjusting the quantization according to the data or to the model loss seems mandatory to enable a high accuracy in the context of quantized neural networks. This work presents Histogram-Equalized Quantization (HEQ), an adaptive framework for linear symmetric quantization. HEQ automatically adapts the quantization thresholds using a unique step size optimization. We empirically show that HEQ achieves state-of-the-art performances on CIFAR-10. Experiments on the STL-10 dataset even show that HEQ enables a proper training of our proposed logic-gated (OR, MUX) residual networks with a higher accuracy at a lower hardware complexity than previous work.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。