AI 中文总结
针对芯片制造导致硬件变异影响DNN推理精度的问题，提出硬件感知软件训练（HCST）方法，以恢复精度并实现高推理精度。
AI 中文摘要
深度神经网络（DNN）已广泛应用于各行业，人们正在探讨专用芯片以实现更低功耗与更高吞吐量。芯片制造过程中产生的硬件变异是影响推理精度的主要原因。本文提出硬件感知软件训练（HCST）方法，该方法可在硬件变异影响下仍实现高推理精度。
英文摘要
Deep neural network (DNN) has been widely applied in various industries. Specialized chips are being discussed for the purpose of achieving lower power consumption with higher throughput. Hardware variations introduced during the process of chip manufacturing are the main reason for affecting the inference accuracies. In this paper, we propose hardware-conscious software training (HCST) method which enables high inference accuracies even under the influence of hardware variations.
Comments2 pages, 11 figures, 2 tables. Presented at SSDM 2023
Journal refExtended Abstracts of the International Conference on Solid State Devices and Materials (SSDM), Nagoya, Japan, September 2023, pp. 427-428
