Machine Learning Toronto (MLTO)
Details
Registration Instructions
NOTE: Meetup RSVPs are closed - to attend, you must register on the Luma event page.
Agenda
6:00-6:30 PM Arrivals and Networking
6:30-7:00 PM Tech Talk #1
7:00-7:30 PM Tech Talk #2
7:30-8:30 PM Networking & Wrap-up
8:30 PM Tear-down and Departures
Talk #1
Jonathan Watson, Bosun AI Strategies - A Thoughtful Perspective on the AI Boom: Misallocation Rather Than Bubble
A recent macro analysis offers a measured view on the massive investments flowing into artificial intelligence. While acknowledging the genuine progress—strong revenues, rapid growth, and real demand—it suggests we may be witnessing not a classic speculative bubble, but a significant misallocation of capital that could limit AI's full potential.
Drawing from economics, physics, and observations of enterprise adoption, the piece highlights four structural challenges in the current approach to building AI infrastructure. These issues, if unaddressed, might temper some of the current optimism.
For those following AI's development, this perspective is shared in the spirit of constructive discussion. It may prompt useful questions about the path ahead and encourage a balanced view of the opportunities emerging in this fast-evolving field.
Talk #2,
John Wilder, Assistant Teaching Professor and Academic Lead of Applied AI MPS Assistant Teaching Professor at Northeastern University - Comparing human and artificial neural networks
In machine learning, a common way to measure differences is to find the difference between two vectors (L1 or L2 distance) or the angle between two vectors (cosine similarity). How can we compare systems that represent the same information in a different way? For example, when there is no feature correspondence between two vectors or even with the vectors are vastly different sizes, can we determine if they are representing information in a functionally similar way? I will present how I compared the human visual system to an artificial vision system, specifically a convolutional neural network (CNN).
Both the human visual system and CNN process visual input in stages. The goal of this work is to compare the stages of the different systems to one another, but there is no correspondence in the representations of the different stages or the different systems. For example, there are 125 million neurons in the retina alone, while VGG16 has 138 million neurons total. I will demonstrate how to use representational similarity analysis to allow for a comparison of these different representations. While I use this technique to compare the human brain to an artificial neural network, they are broadly applicable to comparing representations from any systems, such as different architectures of artificial neural networks.
About the Speakers
Jonathan M. Watson is a strategist and futurist focused on macro-economics, distributed systems, and infrastructure investment. His work examines how physical constraints—especially latency, energy, and scale—shape which technologies succeed and which capital cycles misfire.
Drawing on a background in global media systems and technology strategy, Watson studies the growing mismatch between hyperscaler investment models and the real requirements of edge and real-time compute. He writes and researches independently, sharing frameworks and lessons learned rather than predictions, with a perspective shaped as much by setbacks as by success.
Dr. John Wilder is the faculty lead for the applied AI program and an assistant teaching professor for the analytics program at College of Professional Studies at Northeastern University. He previously served as a lecturer for Northeastern’s Khoury College of Computer Science. John has a history of working on interdisciplinary research in Cognitive Science, Neuroscience, and Computer Vision. His previous positions include a research associate at the University of Toronto and a postdoctoral fellow at the Centre for Vision Research at York University. John earned a PhD in Cognitive Psychology and an MS in Computer Science from Rutgers University. Outside of academia, he has worked as an independent developer of virtual reality software for optometrists. Other interests include board games, canoeing, camping, and reading.
Many thanks to our supporters and sponsors!
We'd also like to acknowledge our venue partners, Northeastern University, for generously providing their amazing space for the event and PRAKTIKAI for making the event happen.
Speak or Support MLTO!
We are always looking for new speakers and sponsors to share their knowledge and help build the AI community in the GTA! If you are interested please get in touch directly or apply via one of the links below:
- Speaker Application Form: mlto.ca/speak
- Sponsor Inquiry Form: mlto.ca/sponsor
- Volunteer Application Form: mlto.ca/volunteer