IBM Analog Hardware Acceleration Kit: A Flexible and Fast PyTorch Toolkit for Simulating ANN Training and Inference on Resistive Crossbar Arrays

Wednesday, February 24, 2021

IBM Analog Hardware Acceleration Kit:
A Flexible and Fast PyTorch Toolkit for Simulating ANN Training and Inference on Resistive Crossbar Arrays

Memristive crossbar arrays are a promising future technology for accelerating AI workloads, but noise and non-idealities demand for improved algorithmic solutions. We introduce the IBM Analog Hardware Acceleration Kit, a first of a kind open source toolkit to simulate crossbar arrays from within PyTorch, to conveniently estimate the impact of material properties and non-idealities on the accuracy for arbitrary ANNs.

Speaker: Malte Rasch, IBM

Explore

Two ions, made of red spheres, meet and a white sphere is transferred.

What Makes a Good Proton Conductor?

Zach Winn | MIT News

MIT researchers found a way to predict how efficiently materials can transport protons in clean energy devices and other advanced technologies.

Energy Efficient AIMaterialsHigh Performance Computation

New Materials Could Boost the Energy Efficiency of Microelectronics

Adam Zewe | MIT News

By stacking multiple active components based on new materials on the back end of a computer chip, this new approach reduces the amount of energy wasted during computation.

Materials

Illustration shows pairs of electrons in magic-angle twisted trilayer graphene

MIT Physicists Observe Key Evidence of Unconventional Superconductivity in Magic-angle Graphene

The findings could open a route to new forms of higher-temperature superconductors.

Jennifer Chu | MIT News