2020年10月18日，IEEE Fellow，杜克大学电子与计算机工程系教授、IEEE Fellow、美国国家自然科学基金会（NSF）产业-大学可持续智能计算合作研究中心（IUCRC）主任，杜克大学计算进化智能中心（CEI）联合主任陈怡然教授应北京大学微纳电子学系邀请，通过网络在线，做了题为《Hardware/Software Co-design for AI Systems》的精彩学术报告。微纳电子学系多位老师以及研究生、本科生参加了报告会，并与报告人进行了热烈的讨论。

Abstract: The rapid growth of modern neural network (NN) models’ scale generates ever-increasing demands for high computing power of artificial intelligence (AI) systems. Many specialized computing devices have been also deployed in the AI systems, forming a truly application-driven heterogenous computing platform. In this talk, we discuss the importance of hardware/software co-design in AI system designs. We first use resistive memory based NN accelerators to illustrate the design philosophy of AI computing systems, and then present several hardware friendly NN model compression techniques. We also extend our discussions to distributed systems and briefly introduce the automation of co-design flow, e.g., neural architecture search.