上一条： Improving Neural Network Efficiency via Post-Training Quantization With Adaptive Floating-Point

下一条： PIMGCN: A ReRAM-Based PIM Design for Graph Convolutional Network Acceleration