AdAM: Adaptive Fault-Tolerant Approximate Multiplier for Edge DNN Accelerators
In this paper, we propose an architecture of a novel adaptive fault-tolerant approximate multiplier tailored for ASIC-based DNN accelerators.
cs.AI↗
arXiv subjects
Publications and source records attributed to Ahsan Rafiq.
In this paper, we propose an architecture of a novel adaptive fault-tolerant approximate multiplier tailored for ASIC-based DNN accelerators.