Kavli Affiliate: Jing Wang | First 5 Authors: Yifeng Zhai, Bing Li, Bonan Yan, Jing Wang, | Summary: RRAM crossbars have been studied to construct in-memory accelerators for neural network applications due to their in-situ computing capability. However, prior RRAM-based accelerators show efficiency degradation when executing the popular attention models. We observed that the frequent […]
Continue.. STAR: An Efficient Softmax Engine for Attention Model with RRAM Crossbar