Design Multi-Model Accelerators via Automatically Extracting Computational Graph Similarities

In this paper, we present a design methodology for multi-model accelerators considering the similar nodes between the two DNN computational graphs. First, we use a maximal common subgraph search algorithm to automatically extract the similarities between computational graphs. Then, we merge the common subgraphs and map the merged computation graph to hardware. We also present an LSTM-GRU hybrid accelerator without the need for redesigning and redeployment. Experimental results indicate that the hybrid accelerator achieves an average reduction of 42.3 % in the total logical resource consumption compared to the independent LSTM and GRU accelerators.