论文标题
生物医学应用的温度感知的整体3D DNN加速器
Temperature-Aware Monolithic 3D DNN Accelerators for Biomedical Applications
论文作者
论文摘要
在本文中，我们专注于温度感知的单片3D（MONO3D）深神经网络（DNN）推理，用于生物医学应用。我们开发了一个优化器，可在用户定义的性能和热约束下调整加速器的纵横比和足迹，并生成近乎最佳的配置。使用拟议的MONO3D优化器，我们证明了比性能优化加速器的生物医学应用的能源效率高达61％。
In this paper, we focus on temperature-aware Monolithic 3D (Mono3D) deep neural network (DNN) inference accelerators for biomedical applications. We develop an optimizer that tunes aspect ratios and footprint of the accelerator under user-defined performance and thermal constraints, and generates near-optimal configurations. Using the proposed Mono3D optimizer, we demonstrate up to 61% improvement in energy efficiency for biomedical applications over a performance-optimized accelerator.
