Small-world-based Structural Pruning for Efficient FPGA Inference of Deep Neural Networks | IEEE Conference Publication | IEEE Xplore