标题： MAx-DNN：用于高效能DNN硬件加速器的多级算术近似 显示英文标题

标题： MAx-DNN: Multi-Level Arithmetic Approximation for Energy-Efficient DNN Hardware Accelerators

摘要： 如今，深度神经网络（DNN）架构的快速增长已使其成为提供先进机器学习任务的默认方法，具有出色的准确性。 针对低功耗DNN计算，本文研究了DNN工作负载的细粒度错误弹性与硬件近似技术之间的相互作用，以实现更高的能源效率。 利用最先进的ROUP近似乘法器，我们根据我们的层、滤波器和内核级别的方法，系统地探索它们在网络中的细粒度分布，并检查它们对准确性和能耗的影响。 我们在CIFAR-10数据集上使用ResNet-8模型来评估我们的近似方法。 与基线量化模型相比，所提出的解决方案在最多4%的准确率损失下，可获得高达54%的能耗节省，同时与最先进的DNN近似方法相比，它提供了2倍的能耗节省且准确率更好。 显示英文摘要

摘要： Nowadays, the rapid growth of Deep Neural Network (DNN) architectures has established them as the defacto approach for providing advanced Machine Learning tasks with excellent accuracy. Targeting low-power DNN computing, this paper examines the interplay of fine-grained error resilience of DNN workloads in collaboration with hardware approximation techniques, to achieve higher levels of energy efficiency. Utilizing the state-of-the-art ROUP approximate multipliers, we systematically explore their fine-grained distribution across the network according to our layer-, filter-, and kernel-level approaches, and examine their impact on accuracy and energy. We use the ResNet-8 model on the CIFAR-10 dataset to evaluate our approximations. The proposed solution delivers up to 54% energy gains in exchange for up to 4% accuracy loss, compared to the baseline quantized model, while it provides 2x energy gains with better accuracy versus the state-of-the-art DNN approximations.