Graph neural networks (GNNs) have become an increasingly popular research topic. Given
its advantages in exploiting the graph data structure, any graph-related problems can be using
the graph neural networks like medicine discovery, recommending systems, and quantum
chemistry modeling. However, accelerating the training and inferencing of GNNs are limited by
problems like Massive irregular memory access and intensive computation for embedding
transformation. Due to these characteristics of graph neural networks, the commonly used
hardware accelerators (GPUs and CPUs) are not ideal for accelerating GNNs.
In this project, we focus on one of the specific variants of GNN, named Graph Convolutional
Network (GCN). Furthermore, we aim to implement the GCN inference model on FPGA. The
primary approach to implementing this project is Vitis HLS, a high-level synthesis tool that can
convert the C code into RTL code running on FPGA.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
