| （英） |
In recent years, Artificial Intelligence, Deep Learning in particular, has become a hot topic of both research and practical applications. However, its success also highlighted its power hungry nature that makes it difficult or even prohibitive to implement for applications where low power real time operation is critical, such as for example self-driving cars. Emerging analog non-volatile memory devices (PCM, ReRAM, FeFET) have been proposed as a promising solution to reduce power consumption by orders of magnitude compared to conventional digital circuits and bring AI applications from cloud to edge. In this talk, we will review recent progress of emerging non-volatile memories for neuro-inspired AI computing and discuss future challenges. |