Our paper “BoostGCN: A Framework for Optimizing GCN Inference on FPGA” has been accepted as full paper at the 29th IEEE International Symposium On Field-Programmable Custom Computing Machines (FCCM ‘21). This paper describes a framework that generates accelerators for various GCN models, and shows significant speedup compared with state-of-the-art implementations on CPU, GPU and FPGA.
