Efficient Message Passing Architecture for GCN Training on HBM-based FPGAs with Orthogonal Topology On-Chip Networks.
Qizhe WuLetian ZhaoYuchen GuiHuawen LiangXiaotian WangXi JinPublished in: FPGA (2024)
Keyphrases
- message passing
- belief propagation
- shared memory
- interconnection networks
- markov random field
- distributed systems
- multithreading
- distributed shared memory
- approximate inference
- reconfigurable hardware
- low cost
- probabilistic inference
- inference in graphical models
- sum product algorithm
- preprocessing
- vlsi implementation
- parallel architectures
- graphical models
- image processing