Technical Resources

Combining accuracy and plasticity in convolutional neural networks with ReRAM

Proposing a hardware solution that combines the benefits of artificial and spiking neural networks (IEEE Journal on Exploratory Solid-State Computational Devices and Circuits)

A Comprehensive Oxide-Based ReRAM TCAD Model with Experimental Verification

Detailing new predictive and physics-based TCAD simulations for modeling Oxide-based ReRAM (Weebit and Silvaco, International Memory Workshop 2021)

Fully-Integrated Spiking Neural Network (SNN) using RRAM as Synaptic Device

The first complete integration of a fully connected SNN with synaptic weights implemented using SiOx based resistive memories (Weebit and CEA-Leti, AICAS 2020)

SiOX-Based ReRAM Memory Window Optimization

Tuning the Initial Resistance to optimize the memory window – with successful use in an SNN architecture for handwritten digit classification (E-MRS Fall Meeting 2019)

Making SiOx ReRAM a Cost-effective Embedded Memory

Commercializing ReRAM designs requires the ability to understand and control important factors affecting ReRAM performance and reliability (Flash Memory Summit 2019)

Implementing Neural Network Synapses with ReRAM

How ReRAM can be used to implement analog accelerators for common deep learning neural networks and brain-inspired spiking neural networks (Flash Memory Summit 2019)