Performance of Graph Neural Networks for Point Cloud Applications

被引:0
|
作者
Parikh, Dhruv [1 ]
Zhang, Bingyi [1 ]
Kannan, Rajgopal [2 ]
Prasanna, Viktor [1 ]
Busart, Carl [2 ]
机构
[1] Univ Southern Calif, Los Angeles, CA 90007 USA
[2] DEVCOM US Army Res Lab, Adelphi, MD USA
关键词
Graph neural network; point cloud; k-nearest neighbors; dynamic graph construction; performance profiling;
D O I
10.1109/HPEC58863.2023.10363595
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Graph Neural Networks (GNNs) have gained significant momentum recently due to their capability to learn on unstructured graph data. Dynamic GNNs (DGNNs) are the current state-of-the-art for point cloud applications; such applications (viz. autonomous driving) require real-time processing at the edge with tight latency and memory constraints. Conducting performance analysis on such DGNNs, thus, becomes a crucial task to evaluate network suitability. This paper presents a profiling analysis of EdgeConv-based DGNNs applied to point cloud inputs. We assess their inference performance in terms of end-to-end latency and memory consumption on state-of-the-art CPU and GPU platforms. The EdgeConv layer has two stages: (1) dynamic graph generation using k-Nearest Neighbors (kNN) and, (2) node feature updation. The addition of dynamic graph generation via kNN in each (EdgeConv) layer enhances network performance compared to networks that work with the same static graph in each layer; such performance enhancement comes, however, at the added computational cost associated with the dynamic graph generation stage (via kNN algorithm). Understanding its costs is essential for identifying the performance bottleneck and exploring potential avenues for hardware acceleration. To this end, this paper aims to shed light on the performance characteristics of EdgeConv-based DGNNs for point cloud inputs. Our performance analysis on a state-of-the-art EdgeConv network for classification shows that the dynamic graph construction via kNN takes up upwards of 95% of network latency on the GPU and almost 90% on the CPU. Moreover, we propose a quasi-Dynamic Graph Neural Network (qDGNN) that halts dynamic graph updates after a specific depth within the network to significantly reduce the latency on both CPU and GPU whilst matching the original networks inference accuracy.
引用
收藏
页数:7
相关论文
共 50 条
  • [41] Point cloud occlusion recovery with shallow feedforward neural networks
    Barazzetti, Luigi
    ADVANCED ENGINEERING INFORMATICS, 2018, 38 : 605 - 619
  • [42] Leveraging Graph Neural Networks for SLA Violation Prediction in Cloud Computing
    Maroudis, Angelos-Christos
    Theodoropoulos, Theodoros
    Violos, John
    Leivadeas, Aris
    Tserpes, Konstantinos
    IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, 2024, 21 (01): : 605 - 620
  • [43] Modeling TCP Performance using Graph Neural Networks
    Jaeger, Benedikt
    Helm, Max
    Schwegmann, Lars
    Carle, Georg
    PROCEEDINGS OF THE 1ST INTERNATIONAL WORKSHOP ON GRAPH NEURAL NETWORKING, GNNET 2022, 2022, : 18 - 23
  • [44] xNet: Modeling Network Performance With Graph Neural Networks
    Huang, Sijiang
    Wei, Yunze
    Peng, Lingfeng
    Wang, Mowei
    Hui, Linbo
    Liu, Peng
    Du, Zongpeng
    Liu, Zhenhua
    Cui, Yong
    IEEE-ACM TRANSACTIONS ON NETWORKING, 2024, 32 (02) : 1753 - 1767
  • [45] Neural networks applications in economics: A statistical point of view
    Lauro, NC
    Davino, C
    Vistocco, D
    NEURAL NETS - WIRN VIETRI-99, 1999, : 357 - 375
  • [46] Clinical applications of graph neural networks in computational histopathology: A review
    Meng, Xiangyan
    Zou, Tonghui
    COMPUTERS IN BIOLOGY AND MEDICINE, 2023, 164
  • [47] A Survey on Privacy in Graph Neural Networks: Attacks, Preservation, and Applications
    Zhang, Yi
    Zhao, Yuying
    Li, Zhaoqing
    Cheng, Xueqi
    Wang, Yu
    Kotevska, Olivera
    Yu, Philip S.
    Derr, Tyler
    IEEE Transactions on Knowledge and Data Engineering, 2024, 36 (12) : 7497 - 7515
  • [48] A Survey on Graph Neural Networks and its Applications in Various Domains
    Tejaswini R. Murgod
    P. Srihith Reddy
    Shamitha Gaddam
    S. Meenakshi Sundaram
    C. Anitha
    SN Computer Science, 6 (1)
  • [49] The Impact of Global Structural Information in Graph Neural Networks Applications
    Buffelli, Davide
    Vandin, Fabio
    DATA, 2022, 7 (01)
  • [50] Graph Neural Networks: A Bibliometric Mapping of the Research Landscape and Applications
    da Silva, Annielle Mendes Brito
    Ferreira, Natiele Carla da Silva
    Braga, Luiza Amara Maciel
    Mota, Fabio Batista
    Maricato, Victor
    Alves, Luiz Anastacio
    Information (Switzerland), 2024, 15 (10)