A Riemannian Geometry Perspective on Foundation Models
Oden Institute for Computational Engineering and Sciences, UT Austin
Check the official event for registration, eligibility and attendance details.
Abstract
Oden Institute Seminar by Rex Ying (Yale University) on how non-Euclidean geometries, particularly hyperbolic geometry, can enhance foundation models by better capturing hierarchies and symmetries in real-world data, with applications across Transformers, language model training, multimodal systems, and recommender systems.
Related Seminars
Learning generative dynamical systems models from multi-modal and multi-animal neuro-data
More on foundation models
On finding what you’re (not) looking for: prospects and challenges for AI-driven discovery
More on machine learning
Llama 3.1 Paper: The Llama Family of Models
More on foundation models