Development and Integration Strategies for Non-volatile Resistive Memory in the Context of Neuromorphic Computing and AI Hardware

Neuromorphic and AI computing hardware is trending away from traditional von Neumann computational architectures. This transition is opening the door to a wide range of novel devices and integration solutions. Over the past 10 years, my research group has focused on fabrication and integration strategies for CMOS-compatible, non-volatile, memory devices (aka: memristors). These memristive devices have the potential to act as neuronal synapses in neural networks, but can also function as tunable elements in array-based accelerators. Introducing unique materials and novel memristive devices into the traditional CMOS fabrication process presents many challenges, from both the process integration and packaging standpoint. In this presentation I will discuss integration strategies that we have used to develop novel hardware solutions and translation of laboratory-scale demonstrations towards integration on 300mm wafer platforms utilizing the Albany Nanotech 300mm foundry.

Speaker(s): Dr. Nate Cady,

Bldg: CNSE, CESTM Audirotium, Albany, New York, United States, Virtual: https://events.vtools.ieee.org/m/316254