Nengo Summer School
Nengo with Neuromorphic Hardware:
The Centre for Theoretical Neuroscience at the University of Waterloo is excited to announce our 7th annual Nengo summer school on large-scale brain modelling and neuromorphic computing. As a recurring special feature of the school, participants will be given on-site access to Loihi, a neuromorphic research chip developed by Intel Labs. Other neuromorphic devices likely to be available at the summer school include Stanford’s mixed digital/analog chip Braindrop, the SpiNNaker chip, and a variety of FPGAs.
In addition to teaching participants to run applications on neuromorphic hardware using the Nengo simulation package, this two-week school will teach participants to use Nengo to build state-of-the-art cognitive and neural models. More generally, Nengo provides users with a versatile and powerful environment for designing cognitive and neural systems, and has been used to build what is currently the world’s largest functional brain model, Spaun, which includes spiking deep learning, reinforcement learning, adaptive motor control, and cognitive control networks.
We welcome applications from all interested graduate students, postdocs, professors, and industry professionals. No specific training in the use of modelling software is required, but we encourage applications from active researchers with a relevant background in psychology, neuroscience, cognitive science, computer science, robotics, neuromorphic engineering, or a related field. More information about Nengo, the Neural Engineering Framework, and Spaun can be found at https://www.nengo.ai/. Many of the concepts to be covered are in this book.
Watch the videos below for a quick look at previous versions of the summer school!
Purpose: The main goal of the summer school is to have participants learn to build state-of-the-art neural or cognitive models using Nengo. Participants are encouraged to bring their own ideas for projects, which may focus on testing hypotheses, modelling neural or cognitive data, implementing specific behavioural functions with neurons, expanding past models, or provide a proof-of-concept of various neural mechanisms. Projects can be focused on software, hardware, or a combination of both.
Approach: Discussions during the school will help shape projects, develop new collaborations, and generate new ideas for projects. Participants will have the opportunity to learn how to:
- use Nengo to program neuromorphic hardware systems, including Loihi!
- integrate machine learning methods into biologically oriented models
- interface Nengo with cameras and robotic systems of various kinds
- build perceptual, motor, and cognitive models with spiking neurons
- simulate large-scale models at a variety of levels of detail concurrently
- model anatomical and electrophysiological data
- model cognitive and behavioural data with a detailed neural model
- use a variety of single cell models within a large-scale model
- use your favorite simulator, e.g. NEST, Brian, Neuron, etc.
- implement modern non-linear control methods in a neural model
- and much more …
Format: The first few days of the summer school will be be devoted to in-depth tutorials on using Nengo for neural and cognitive modelling. The rest of summer school will be devoted primarily to project work, with a variety of faculty guest lectures and optional tutorials scheduled throughout. A project demonstration and evening banquet will be held on the last day, with a variety of prizes for strong projects! The complete schedule and list of guest lecturers will be announced at a later date.
Dates: June 7th - June 19th, 2020
Location: University of Waterloo, Ontario, Canada (map)
Cost: A registration fee is required of all participants (USD $450 for academic participants; USD $900 for industry participants).
This fee includes the cost of breakfast, lunch, and snacks during each weekday of the school, along with a summer school T-shirt! Some evening meals will also be covered. Participants are responsible for covering the cost of their travel, accommodations, and recreational activities.
Travel: For most participants, arrival through Toronto’s Pearson International Airport is the most convenient option. The University of Waterloo is approximately a one hour drive away, and a number of shuttle services are available to take you directly to the University from the airport. A list of general travel options can be found here
Accommodations: On-campus accommodations will be made available to all confirmed participants. The summer school rates for these accommodations will be announced at a later date.
Applications: To apply, you will need to submit a CV and a brief proposal for a project. At the summer school you’ll be able to make use of neuromorphic hardware, event-based vision sensors, simple humanoid and non-humanoid robotic devices, and a variety of large-scale cognitive modelling techniques. If applicable, please briefly indicate how you might leverage this infrastructure in your project. The application form is available here.
Application deadline: Applications are due on February 15th, 2020.
Contact information: For any questions, please contact Peter Blouw.