Deep learning models are being deployed in many mobile intelligent applications. End-side services, such as intelligent personal assistants, autonomous cars, and smart home services often employ either simple local models on the mobile or complex remote models on the cloud. However, recent studies have shown that partitioning the DNN computations between the mobile and cloud can increase the latency and energy efficiencies. In this paper, we propose an efficient, adaptive, and practical engine, JointDNN, for collaborative computation between a mobile device and cloud for DNNs in both inference and training phase. JointDNN not only provides an energy and performance efficient method of querying DNNs for the mobile side but also benefits the cloud server by reducing the amount of its workload and communications compared to the cloud-only approach. Given the DNN architecture, we investigate the efficiency of processing some layers on the mobile device and some layers on the cloud server. We provide optimization formulations at layer granularity for forward- and backward-propagations in DNNs, which can adapt to mobile battery limitations and cloud server load constraints and quality of service. JointDNN achieves up to 18 and 32 times reductions on the latency and mobile energy consumption of querying DNNs compared to the status-quo approaches, respectively.

中文翻译：

---

JointDNN：智能移动云计算服务的高效训练和推理引擎

深度学习模型正在许多移动智能应用程序中部署。终端服务，例如智能个人助理、自动驾驶汽车和智能家居服务，通常要么在移动端使用简单的本地模型，要么在云端使用复杂的远程模型。然而，最近的研究表明，在移动和云之间划分 DNN 计算可以增加延迟和能源效率。在本文中，我们提出了一种高效、自适应且实用的引擎 JointDNN，用于移动设备和云之间的协作计算，用于推理和训练阶段的 DNN。JointDNN 不仅为移动端查询 DNN 提供了一种能源和性能高效的方法，而且与纯云方法相比，通过减少其工作负载和通信量，使云服务器受益。鉴于 DNN 架构，我们研究了在移动设备上处理某些层和在云服务器上处理某些层的效率。我们为 DNN 中的前向和后向传播提供层粒度的优化公式，可以适应移动电池限制和云服务器负载限制和服务质量。与现状方法相比，JointDNN 查询 DNN 的延迟和移动能耗分别降低了 18 倍和 32 倍。可以适应手机电池限制和云服务器负载限制和服务质量。与现状方法相比，JointDNN 查询 DNN 的延迟和移动能耗分别降低了 18 倍和 32 倍。可以适应手机电池限制和云服务器负载限制和服务质量。与现状方法相比，JointDNN 查询 DNN 的延迟和移动能耗分别降低了 18 倍和 32 倍。