SimBrain – Neuromorphic Simulation Framework

Electrical Engineering

Team 35

Lucien Roberson, Chiraz Barhoumi, Sabryna Henze, Bryan Matos, Dennis Hoang

Summary

Compares traditional and spiking neural networks under realistic hardware conditions. Using IBM’s AIHWKIT simulator, the project measures how analog device imperfections such as RRAM read noise and conductance drift affect model performance. The goal is to quantify the real-world tradeoffs of running neural networks on energy-efficient neuromorphic hardware using a controlled comparison framework.

Demo Day Zoom link: https://asu.zoom.us/j/2984187484

Video

Research poster

Sponsor

Advisor

Portrait of Soyed Tuhin AHMED

Soyed Tuhin AHMED

Postdoctoral Research Scholar

[email protected]