Assuming hardware is the major constraint for enabling real-time mobile intelligence, the industry has mainly dedicated their efforts to developing specialized hardware accelerators for machine learning and inference. This article challenges the assumption. By drawing on a recent real-time AI optimization framework CoCoPIE, it maintains that with effective compression-compiler co-design, it is possible to enable real-time artificial intelligence on mainstream end devices without special hardware. CoCoPIE is a software framework that holds numerous records on mobile AI: the first framework that supports all main kinds of DNNs, from CNNs to RNNs, transformer, language models, and so on; the fastest DNN pruning and acceleration framework, up to 180X faster compared with current DNN pruning on other frameworks such as TensorFlow-Lite; making many representative AI applications able to run in real-time on off-the-shelf mobile devices that have been previously regarded possible only with special hardware support; making off-the-shelf mobile devices outperform a number of representative ASIC and FPGA solutions in terms of energy efficiency and/or performance.

中文翻译：

---

CoCoPIE：让移动AI像PIE一样甜蜜——压缩编译协同设计大有作为

假设硬件是实现实时移动智能的主要制约因素，业界主要致力于开发用于机器学习和推理的专用硬件加速器。这篇文章挑战了这个假设。通过借鉴最新的实时人工智能优化框架 CoCoPIE，它认为通过有效的压缩-编译器协同设计，可以在没有特殊硬件的情况下在主流终端设备上启用实时人工智能。CoCoPIE 是一个在移动 AI 上拥有众多记录的软件框架：第一个支持所有主要 DNN 类型的框架，从 CNNs 到 RNNs、transformer、语言模型等等；最快的 DNN 修剪和加速框架，与当前在其他框架（如 TensorFlow-Lite）上的 DNN 修剪相比，速度提高了 180 倍；使许多具有代表性的 AI 应用程序能够在现成的移动设备上实时运行，而这些应用程序以前被认为只有在特殊的硬件支持下才有可能；使现成的移动设备在能效和/或性能方面优于许多具有代表性的 ASIC 和 FPGA 解决方案。