An FPGA-based model suitable for evolution and development of spiking neural networks

AbstractWe propose a digital neuron model suitable for evolving and growing heterogeneous spiking neural networks on FPGAs using a piecewise linear approximation of the Quadratic Integrate and Fire (QIF) model. A network of 161 neurons and 1610 synapses with 4210 times realtime neuron simulation speed was simulated and synthesized for a Virtex-5 chip.

Download publication

Related Resources

See what’s new.



CLIP-Forge: Towards Zero-Shot Text-to-Shape Generation

Generating shapes using natural language can enable new ways of…



Exploring the Collective Categorization of Biological Information for Biomimetic Design

Categorizing biological information can be subjective and ambiguous,…



Considering Multiscale Scenes to Elucidate Problems Encumbering 3D Intellection and Navigation

Virtual 3D environments have become pervasive tools in a number of…



Survey-Based Simulation of User Satisfaction for Generative Design in Architecture

This paper describes a novel humanist approach to generative design…

Get in touch

Something pique your interest? Get in touch if you’d like to learn more about Autodesk Research, our projects, people, and potential collaboration opportunities.

Contact us