Novel Device and Materials in Emerging Memory for Neuromorphic Computing

Explore novel device technologies and materials for emerging memory in neuromorphic computing in this tinyML talk. Delve into the development of new hardware solutions to overcome Moore's law scaling limits and enable efficient data-driven computing with high reliability. Learn about self-rectified memory, two-dimensional memory, self-aligned helical-structured materials, and thermal stability investigations for next-generation memory. Discover how these advancements contribute to neuromorphic computational applications and address challenges like sneak path current in array applications. Examine the fabrication process of bilayer selectorless RRAM, array size considerations, and read margin improvements. Investigate the detection of resistive switching gaps, mechanisms behind bilayer selectorless RRAM, and its benchmarking. Gain insights into synaptic behavior for neuromorphic computing, memory window effects with radiation, and temperature impacts during training and inference processes.

Intro
Why Memory is Important?
Memory Hierarchy
Emerging Memory and Applications
Memory Categories
Resistive Random Access Memory and Crossbar A
Problem: Sneak Path Current (SPC) in Array Applicatio
Solution for Sneak Path Current
Fabrication for Bilayer Selectorless RRAM
Array Size and Read Margin 50
Future Path for High Storage Class Memory
Can the resistive switching gap be detected?
Can the switching gap be detected?
Mechanism for Bilayer Selectorless RRA
Benchmark of Bilayer Selectorless RRAM
Synaptic Behavior for Neuromorphic Computi
Memory Window with Radiaiton Effect
Temperature Effect during Training and Inference Pro
Silver Strategic Partners