Toward HDL Extensions for Rapid AI/ML Accelerator Generation | IEEE Conference Publication | IEEE Xplore