MaGNAS：面向异构 MPSoC 部署的映射感知图神经架构搜索框架
MaGNAS: A Mapping-Aware Graph Neural Architecture Search Framework for Heterogeneous MPSoC Deployment
AI总结：
提出 MaGNAS 映射感知图神经架构搜索框架，通过双层进化搜索联合优化视觉 GNN 架构与异构 MPSoC 映射，在 Xavier AGX 上显著降低延迟和能耗且准确率损失极小。
摘要：
图神经网络（GNN）因其在建模图像帧各部分之间结构和上下文关系方面的内在能力，在基于视觉的应用中日益流行。另一方面，边缘端深度视觉应用的兴起得益于异构多处理器片上系统（MPSoC）的最新进展，这类系统能够在实时且严格的执行要求下实现推理。进一步而言，用于视觉应用的 GNN 也必须满足相同的执行要求。然而，与典型的深度神经网络不同，图学习操作的不规则流程给在此类异构 MPSoC 平台上运行 GNN 带来了挑战。本文提出一种新颖的统一设计—映射方法，用于在异构 MPSoC 平台上高效处理视觉 GNN 工作负载。具体而言，我们开发了 MaGNAS，这是一个映射感知的图神经架构搜索框架。MaGNAS 提出了一个 GNN 架构设计空间，并结合异构 SoC 上的预期映射选项，以识别最大化设备端资源效率的模型架构。为实现这一目标，MaGNAS 采用双层进化搜索来识别最佳 GNN 与映射配对，从而获得最佳性能权衡。通过设计一个源自近期 Vision GNN（ViG）架构的超网，我们在四个（04）最先进的视觉数据集上开展实验，同时使用（i）真实硬件 SoC 平台（NVIDIA Xavier AGX）和（ii）面向 DNN 加速器的性能/成本模型模拟器。实验结果表明，相较于仅使用 GPU 的部署，MaGNAS 在 Xavier MPSoC 上执行多个视觉数据集时可提供 1.57 倍延迟加速，并实现 3.38 倍的能效提升，同时相对基线仅保持平均 0.11% 的准确率下降。
英文摘要：
Graph Neural Networks (GNNs) are becoming increasingly popular for vision-based applications due to their intrinsic capacity in modeling structural and contextual relations between various parts of an image frame. On another front, the rising popularity of deep vision-based applications at the edge has been facilitated by the recent advancements in heterogeneous multi-processor Systems on Chips (MPSoCs) that enable inference under real-time, stringent execution requirements. By extension, GNNs employed for vision-based applications must adhere to the same execution requirements. Yet contrary to typical deep neural networks, the irregular flow of graph learning operations poses a challenge to running GNNs on such heterogeneous MPSoC platforms. In this paper, we propose a novel unified design-mapping approach for efficient processing of vision GNN workloads on heterogeneous MPSoC platforms. Particularly, we develop MaGNAS, a mapping-aware Graph Neural Architecture Search framework. MaGNAS proposes a GNN architectural design space coupled with prospective mapping options on a heterogeneous SoC to identify model architectures that maximize on-device resource efficiency. To achieve this, MaGNAS employs a two-tier evolutionary search to identify optimal GNNs and mapping pairings that yield the best performance trade-offs. Through designing a supernet derived from the recent Vision GNN (ViG) architecture, we conducted experiments on four (04) state-of-the-art vision datasets using both (i) a real hardware SoC platform (NVIDIA Xavier AGX) and (ii) a performance/cost model simulator for DNN accelerators. Our experimental results demonstrate that MaGNAS is able to provide 1.57x latency speedup and is 3.38x more energy-efficient for several vision datasets executed on the Xavier MPSoC vs. the GPU-only deployment while sustaining an average 0.11% accuracy reduction from the baseline.
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。
