selected publications conference paper Demonstration of a Distributed Accelerator Framework for Energy-efficient ML Processing 2021