This article presents an approximate signed digit representation (ASD) which quantizes the weights of convolutional neural networks (CNNs) in order to make multiplier-less CNNs without performing any retraining process. Unlike the existing methods that necessitate retraining for weight quantization, the proposed method directly converts full-precision weights of CNN models into low-precision ones, attaining accuracy comparable to that of full-precision models on the Image classification tasks without going through retraining. Therefore, it is effective in saving the retraining time as well as the related computational cost. As the proposed method simplifies the weights to have up to two non-zero digits, multiplication can be realized with only add and shift operations, resulting in a speed-up of inference time and a reduction of energy consumption and hardware complexity. Experiments conducted for famous CNN architectures, such as AlexNet, VGG-16, ResNet-18 and SqueezeNet, show that the proposed method reduces the model size by 73% at the cost of a little increase of error rate, which ranges from 0.09% to 1.5% on ImageNet dataset. Compared to the previous architecture built with multipliers, the proposed multiplier-less convolution architecture reduces the critical-path delay by 52% and mitigates the hardware complexity and power consumption by more than 50%.

中文翻译：

---

无乘数卷积神经网络的无再训练权重量化

本文提出了一种近似有符号数字表示 (ASD)，它量化了卷积神经网络 (CNN) 的权重，以便在不执行任何再训练过程的情况下制作无乘数的 CNN。与现有方法需要重新训练权重量化不同，所提出的方法直接将 CNN 模型的全精度权重转换为低精度权重，在图像分类任务上获得与全精度模型相当的精度，而无需经过重新训练。因此，它有效地节省了再训练时间以及相关的计算成本。由于所提出的方法将权重简化为最多两个非零数字，因此乘法可以仅通过加法和移位运算来实现，从而加快推理时间并降低能耗和硬件复杂性。对著名的 CNN 架构（如 AlexNet、VGG-16、ResNet-18 和 SqueezeNet）进行的实验表明，所提出的方法将模型大小减少了 73%，但错误率略有增加，错误率从 0.09% 到ImageNet 数据集上的 1.5%。与之前使用乘法器构建的架构相比，所提出的无乘法器卷积架构将关键路径延迟降低了 52%，并将硬件复杂性和功耗降低了 50% 以上。ImageNet 数据集上的 09% 到 1.5%。与之前使用乘法器构建的架构相比，所提出的无乘法器卷积架构将关键路径延迟降低了 52%，并将硬件复杂性和功耗降低了 50% 以上。ImageNet 数据集上的 09% 到 1.5%。与之前使用乘法器构建的架构相比，所提出的无乘法器卷积架构将关键路径延迟降低了 52%，并将硬件复杂性和功耗降低了 50% 以上。