ABSTRACT
This chapter explores the current trend and needs for deep neural networks (DNN) processing in realtime in resource-constrained hardware like IoT and discusses the different challenges and opportunities. The chapter summarizes the latest developments in accelerating DNN. The various architectures for DNN execution that present very novel solutions in terms of computing units, dataflow optimization, network topologies which are targeted and architectures for emerging applications are discussed with a comparative study. A vision of the future trend of AI chip designs is also presented.
