arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.02968cs.LGcs.AIcs.NI
收藏

一种基于软Actor-Critic的自适应设备-边缘协同推理框架

An Adaptive Device-Edge Co-Inference Framework Based on Soft Actor-Critic

University of Houston(休斯顿大学);Beijing University of Posts and Telecommunications(北京邮电大学)

Tao Niu, Yinglei Teng, Zhu Han, Panpan Zou

摘要:

近年来,深度神经网络(DNN)由于其优越的特征提取性能,在计算机视觉(CV)和自然语言处理(NLP)等多个领域的应用非常突出。然而,高维参数模型和大规模数学计算限制了执行效率,特别是对于物联网(IoT)设备。与以往仅云/边缘模式给上行通信带来巨大压力、仅设备模式承担难以承受的计算强度不同,DNN模型突出了设备与边缘之间的协同计算,可以在通信负载和执行精度之间实现良好的平衡。具体来说,提出了一个系统的按需协同推理框架来利用多分支结构,其中预训练的Alexnet通过\emph{early-exit}进行适当的大小调整,并在中间DNN层进行分区。实施整数量化以进一步压缩传输比特。因此,我们建立了一个新的深度强化学习(DRL)优化器-离散软演员评论(SAC-d),它通过软策略迭代生成\emph{exit point},\emph{partition point}和\emph{compressing bits}。基于延迟和准确度感知的奖励设计,这种优化器可以很好地适应动态无线信道和任意CPU处理等复杂环境,并能够支持5G URLLC。在Raspberry Pi 4和PC上的实际实验表明了该解决方案的优越性。

英文摘要:

Recently, the applications of deep neural network (DNN) have been very prominent in many fields such as computer vision (CV) and natural language processing (NLP) due to its superior feature extraction performance. However, the high-dimension parameter model and large-scale mathematical calculation restrict the execution efficiency, especially for Internet of Things (IoT) devices. Different from the previous cloud/edge-only pattern that brings huge pressure for uplink communication and device-only fashion that undertakes unaffordable calculation strength, we highlight the collaborative computation between the device and edge for DNN models, which can achieve a good balance between the communication load and execution accuracy. Specifically, a systematic on-demand co-inference framework is proposed to exploit the multi-branch structure, in which the pre-trained Alexnet is right-sized through \emph{early-exit} and partitioned at an intermediate DNN layer. The integer quantization is enforced to further compress transmission bits. As a result, we establish a new Deep Reinforcement Learning (DRL) optimizer-Soft Actor Critic for discrete (SAC-d), which generates the \emph{exit point}, \emph{partition point}, and \emph{compressing bits} by soft policy iterations. Based on the latency and accuracy aware reward design, such an optimizer can well adapt to the complex environment like dynamic wireless channel and arbitrary CPU processing, and is capable of supporting the 5G URLLC. Real-world experiment on Raspberry Pi 4 and PC shows the outperformance of the proposed solution.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑