一句话：提出 Hidet 的任务映射编程范式，以简化复杂加速器上深度学习张量程序的开发。
随着深度学习模型在云端服务和边缘设备上广泛部署，降低模型推理延迟对高效服务至关重要。然而，现代加速器（如 NVIDIA GPU、Google TPU）高度复杂，且深度学习算子数量快速增长，导致为这些算子开发高效张量程序非常困难。Hidet 提出任务映射（task-mapping）编程范式，面向深度学习张量程序，试图应对上述开发复杂性。给出的摘要未展开具体实现、编程抽象及实验评估结果。
As deep learning models nowadays are widely adopted by both cloud services and edge devices, reducing the latency of deep learning model inferences becomes crucial to provide efficient model serving. However, it is challenging to develop efficient tensor programs for deep learning operators due to the high complexity of modern accelerators (e.g., NVIDIA GPUs and Google TPUs) and the rapidly growing number of operators.
@inproceedings{paperbot10023,
title = {Hidet: Task-Mapping Programming Paradigm for Deep Learning Tensor Programs},
author = {Yaoyao Ding and Cody Hao Yu and Bojian Zheng and Yizhi Liu and Yida Wang and Gennady Pekhimenko},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3575693.3575702}
}