A High Level Synthesis Framework for Spiking Neural Networks

A High Level Synthesis Framework for Spiking Neural Networks

Internship Description

Nowadays, AI is defined as the new electricity powered by the deep neural networks (DNN). Even though DNNs resembles the human brain, it is considered as the first generation of the neural networks. On the other hand, there are other types of neural networks closer to the human brain in terms of the behaviour. Spiking Neural Networks (SNNs) can be considered in this class of neural networks and called as second generation neural networks. Even though the neuron of SNNs can resemble the human brain more closely, their hardware realizations become more costly and complicated. Therefore, an effiicient high level synthesis framework could be very helpful for the researchers working in this area.

Recommended Student Academic & Research Background: ​Logic circuits, basic HDL coding, programming, VHDL, Verilog, CS
​​

Deliverables/Expectations

Objectives: To provide a toolset for high level synthesis of SNNs that synthesizes the corresponding HDL of the provided SNN. Tools to learn: Python, HDL, FPGA, Cadence Tools

Faculty Name

Khaled Nabil Salama

Field of Study

Computer Science, Neuroscience, Electrical Engineering ​