Canadian Neuromorphic Engineering Workshop

Large-Scale SNNs on Neuromorphic Hardware

Real-time simulation of biologically-representative spiking neural networks at scale, and the practical challenges of mapping them onto neuromorphic hardware.

Event-Driven Sensing for Edge Vision

Event-based sensor fusion for computer vision at the edge — covering in-sensor and near-sensor computing paradigms for low-power deployments.

Tutorial: Beyond von Neumann

Hands-on exploration of neuromorphic architectures designed to overcome the memory-bandwidth bottleneck.

Visual Processing & Sensory Coding

How visual information is transformed across the stages of the visual hierarchy to form everyday perception — bridging large-scale spiking network models, rate-based models, and machine learning.

Neuroprosthetic Vision Restoration

Applying computational neuroscience to design stimulation protocols for future sensory prosthetic systems, from biological modelling to practical implant design.

Tutorial: Quantified Self & Teaching with Neurons

An hands-on session exploring self-tracking data through a neuroscience lens, and pedagogical approaches for teaching computational concepts with spiking models.

“Neuromorphic control through the eyes of hybrid systems”

An introduction to neuromorphic approaches to control engineering. We discuss how theoretical tools from hybrid dynamical systems and event-triggered control and estimation can be used to model, analyze and design spiking controllers with guarantees.

Spiking Controllers for Different Control Problems

How spiking control architectures can address classical control objectives such as stabilization, regulation, and rhythmic control — and the advantages they may offer compared to classical techniques. Case studies include neuromorphic control for nuclear fusion plasma fueling using ice pellets.

Tutorial: From Event-Based Control to Spiking Control Systems

A hands-on session bridging classical control questions and tools with neuromorphic design — covering how to model and analyze closed-loop systems with spikes by exploiting their hybrid and event-based nature.

Insect Navigation & Vision

Key differences from mammalian vision, efficient navigation strategies, and their relevance to low-power neuromorphic computation.

Intro to Neuromorphic Engineering

A broad overview of the landscape — analogue vs. digital approaches and example systems.

Tutorial: GeNN / mlGeNN

GPU-accelerated spiking neural network simulation with GeNN and its machine learning companion mlGeNN.

What does it really mean to “compute”?

The observation that functional programming (born of the lambda calculus) and imperative programming (born of the Turing machine) are equivalent is foundational to computer science. Yet, there’s a substantial imbalance in the visibility of these two paradigms. Imperative programming is essentially the de facto model of computation in the world today. Through it, we emphasise state and data; computation is a game of manipulating symbols, and the properties of those symbols are important above all else. Meanwhile, functional programming eschews what’s “inside” the symbols. Primitive types are fairly close to vacuous, and there’s little distinction between data and functions between data. From this lens, a different emphasis emerges. I argue that the (real) spirit of computation—composition and abstraction—is more easily seen from this lens, where suggested best practices in imperative approaches become core pillars. This switch in thinking is made rigorous by category theory, and suggests a broader, philosophical shift from noun-based ways of thinking to verb-based ones.

How the Brain Works — and How to Build It

A three-part arc: how the brain computes, how to engineer systems that follow the same principles, and how to put those systems to practical use.

Tutorial: Introduction to Nengo

Hands-on introduction to Nengo for building and simulating large-scale brain-inspired models across multiple hardware backends.

Tutorial: Efficient Representations of Time and Space

How spiking networks encode temporal and spatial information efficiently, with practical exercises building such representations from scratch.

Mammalian Vision & Event-Based Perception

How the mammalian visual system processes the world — from retinal encoding to cortical representations — and how these principles translate into neuromorphic vision systems using dynamic vision sensors.

Attention & Microsaccades in Neuromorphic Systems

Biologically-inspired active vision: how microsaccadic eye movements drive efficient scene exploration, and how to implement event-based attention mechanisms on neuromorphic hardware.

Tutorial: DVS Cameras & SpiNNaker

Hands-on session working with dynamic vision sensors and the SpiNNaker platform — from raw event streams to real-time spiking network deployment for computer vision tasks.

Probabilistic Neuromorphic Programming

A new paradigm for programming neuromorphic systems probabilistically — bridging cognitive architectures, generative models, and Vector Symbolic Algebras for principled uncertainty handling in spiking networks.

Automous Exploration

From planetary robotics to animal exploration: how biologically-inspired representations enable efficient active exploration — drawing on work at NASA Ames and the NRC.

Tutorial: Vector Symbolic Architectures

Hands-on introduction to VSAs as a framework for structured, compositional representations in spiking neural networks — covering binding, bundling, and probabilistic inference in high-dimensional spaces.

Neuromorphic algorithms for scientific computing

Due to increasing power costs, neuromorphic computing is starting to be explored for large-scale scientific computing applications. We will discuss how classic compute algorithms, such as random walk simulations and graph analytics can be accelerated on neuromorphic hardware.

A theoretical framework for neuromorphic computing

We will discuss how neuromorphic algorithms scale compared to their conventional counterparts and what this means in terms of algorithms suitability for neuromorphic implementation.

Tutorial: NeuroFEM

Hands-on introduction to directly solving Finite Element Method (FEM) problems using spiking neural networks.

Visual Selective Attention & New Developments

How the brain selects and prioritises relevant visual information from a complex scene — covering the computational principles of selective attention in primates and recent developments in the field, including figure-ground organisation, border-ownership coding, and motion-driven saliency.

Tutorial: The Mihalas–Niebur Neuron Model

A hands-on introduction to the generalised integrate-and-fire Mihalas–Niebur neuron model — covering its biological motivations, its rich repertoire of spiking behaviours, and how to implement it in practice for large-scale network simulation.

Analog Neuromorphic Computing & Field Programmable Analog Arrays

How floating-gate MOS transistors and large-scale FPAAs enable reconfigurable analog hardware that physically implements neural computation — and why analog is thousands of times more energy-efficient than digital for neuromorphic workloads.

Computing in Memory & Low-Power Neural Systems

The principles of in-memory computing, subthreshold transistor models of biological neurons and synapses, and how mixed-signal IC design bridges the gap between neuroscience and real hardware deployment.

Tutorial: FPAAs for Neuromorphic Engineering

Hands-on introduction to Field Programmable Analog Arrays — from synthesis tools to on-chip learning — showing how to map spiking neural network computations directly onto reconfigurable analog substrates.

Neuromodulation & Neocortical Network Dynamics

How neuromodulatory systems shape the activity of cortical networks — drawing on experimental neurophysiology and large-scale data-driven simulations developed at the Blue Brain Project, EPFL.

Data-Driven In Silico Brain Reconstruction

The principles behind reconstructing brain regions computationally: how experimental data constrains model building, and what large-scale simulations reveal about emergent network behaviour.

Tutorial: Large-Scale Biologically Detailed Network Modelling

Hands-on introduction to building and running biophysically detailed spiking network models, informed by Blue Brain Project tools and methodology.