Energy-efficient DNN Inference on Approximate Accelerators Through Formal Property Exploration

Open in new window