While the role of Deep Neural Networks (DNNs) in a wide range of safety-critical applications is expanding, emerging DNNs experience massive growth in terms of computation power. It raises the necessity of improving the reliability of DNN accelerators yet reducing the computational burden on the hardware platforms, i.e. reducing the energy consumption and execution time as well as increasing the efficiency of DNN accelerators. Therefore, the trade-off between hardware performance, i.e. area, power and delay, and the reliability of the DNN accelerator implementation becomes critical and requires tools for analysis. In this paper, we propose a framework DeepAxe for design space exploration for FPGA-based implementation of DNNs by considering the trilateral impact of applying functional approximation on accuracy, reliability and hardware performance. The framework enables selective approximation of reliability-critical DNNs, providing a set of Pareto-optimal DNN implementation design space points for the target resource utilization requirements. The design flow starts with a pre-trained network in Keras, uses an innovative high-level synthesis environment DeepHLS and results in a set of Pareto-optimal design space points as a guide for the designer. The framework is demonstrated in a case study of custom and state-of-the-art DNNs and datasets.
翻译:虽然深神经网络(DNN)在一系列广泛的安全关键应用中的作用正在扩大,但新兴的DNN网络在计算能力方面经历了巨大的增长,因此有必要提高DNN加速器的可靠性,同时减轻硬件平台的计算负担,即减少能源消耗和执行时间,提高DNN加速器的效率。因此,硬件性能(即面积、功率和延迟)之间的交换以及DNN加速器执行的可靠性变得至关重要,需要分析工具。在本文件中,我们提出一个设计空间探索的DeepAxe框架,用于设计基于FPGA的DNN加速器实施DN的太空探索,考虑对精度、可靠性和硬件性能应用功能近似性的三边影响。该框架使得对可靠性至关重要的DNNN能够有选择地对齐,为目标资源利用要求提供一套Pareto-optimal DNNN执行空间设计空间空间设计点。设计流程从在Keras经过预先培训的网络开始,使用创新的高级合成环境深海HLS和S-Simal-Statal Statimal-Deal-Dedual-Dedual-Degal-Degal-Develal-Develal-Develal-Develal-Stal-Develal-Develal-Develal-Develal-Develal-Stal-Develal-Stal-Deal-Deal-Develal-Develal-st-stal-stal-stal-stal-stal-stal-st-st-stal-stal-st-st-st-st-s-tocal-stal-stal-st-st-st-st-st-st-st-st-st-st-st-s-stal-stal-deal-st-st-s-st-st-st-st-st-st-sal-stal-s-sal-sal-toal-toal-tocal-tocal-tocal-toal-toal-toal-s-s-s-s-s-s-s-s-s-s-todal-todal-toal-s-s-s-s-toal-toal-s</s>