Date: 14 Feb 2024 @3:00PM-04:00PM

Title: Modelling neural probabilistic computation using vector symbolic architectures

Location: Online & DT 2203

Abstract:

Distributed vector representations, and specifically Vector Symbolic Architectures (VSAs), are a key bridging point between connectionist and symbolic representations in cognition.. In this talk we will discuss how bundles of symbols in certain Vector Symbolic Architectures (VSAs) can be understood as defining an object that has a relationship to a probability distribution, and how statements using the Holographic Reduced Representation algebra are analogous to probabilistic statements.. We will discuss how dot-product similarity between vector representations of data can be used in density estimation.. We will further show how populations of spiking neurons can be used to implement probabilistic operations that are useful in building cognitive models.

Speaker: Michael Furlong

Bio:

Michael Furlong is a Research Associate in the Centre for Theoretical Neuroscience at the University of Waterloo, and an Associate Research Officer at the National Research Council of Canada.  His research is at the intersection of probabilistic algorithms and neural computation.. Previously, he worked on automating planetary exploration at NASA Ames’ Intelligent Robotics Group.. He holds a doctorate in Robotics from Carnegie Mellon University, a master’s in Neuroscience from Oxford, and a B.Eng in Electrical and Computer Engineering from Memorial University.

Dr. Furlong is also organizing the Nengo Summer School. Information below:

[Applications can be submitted at this link: https://forms.gle/xhkNENnpPHeBXu5W7]

[All details about this school can be found online at https://www.nengo.ai/summer-school]

The Centre for Theoretical Neuroscience at the University of Waterloo is excited to announce our 9th annual Nengo summer school on large-scale brain modelling and neuromorphic computing. This two-week school will teach participants to use the Nengo simulation package to build state-of-the-art cognitive and neural models to run both in simulation and on neuromorphic hardware. Summer school participants will be given on-site access to neuromorphic hardware and will learn to run high-level applications using Nengo! 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.

For a look at the last in-person summer school, check out this short video: https://youtu.be/vRcN5-0wdQE

We welcome applications from all interested graduate students, postdocs, professors, and industry professionals with a relevant background.