Geometric Algebra Transformers: Revolutionizing Geometric Data with Taco Cohen, Qualcomm AI Research

Join us for an illuminating discussion as Taco Cohen, a renowned machine learning researcher at Qualcomm AI Research, unveils the groundbreaking Geometric Algebra Transformer (GATr) in this captivating video presentation. Dive into the world of geometric data and discover how GATr, a versatile neural network architecture, is changing the game across fields like computer vision, robotics, chemistry, and physics.
🔍 Explore the GATr's ability to handle diverse geometric data types, from points and vectors to transformations and beyond.
🌐 Learn how GATr leverages projective geometric algebra for efficient 16-dimensional vector space representations of common geometric objects and operators.
🔄 Discover GATr's equivariance with respect to E(3), the symmetry group of 3D Euclidean space, and how it impacts its scalability and expressiveness.
📈 Witness the remarkable performance improvements GATr brings to various geometric problems compared to non-geometric baselines.
Taco Cohen's journey, from co-founding Scyfer, a deep active learning company, to his remarkable achievements in the field of machine learning, sets the stage for this insightful presentation. Gain invaluable insights into equivariant networks, geometric deep learning, and more as Taco shares his expertise.
Don't miss out on this opportunity to explore the future of geometric data processing with Taco Cohen and the Geometric Algebra Transformer. Like, subscribe, and hit the notification bell to stay updated on cutting-edge AI research!
This enlightening presentation, delivered at the Intelligent Systems Conference 2023 in Amsterdam, The Netherlands.
🌐 Learn more about Intellisys: saiconference.com/IntelliSys

[@KennethFeur]

Very informative lecture. Thank you! But when you edit a video don't change scenes too frequently please... (also some people watch videos in 2x speed so for them the experience is even worse)

[@eternaldoorman5228]

The idea of a 6G system using AI to optimise transmission power inside a closed space sounds really scary to me.

[@pre-universitygeometricalg5862]

You may be interested in the LinkedIn group that is the source of the YT channel that I administer. The group is (unsurprisingly!) "Pre-University Geometric Algebra", and was founded as a space where Ga students, teachers, and subject-matter experts can network and collaborate on developing instructional materials.