#85 Dr. Petar Veličković (Deepmind) - Categories, Graphs, Reasoning [NEURIPS22 UNPLUGGED]
Machine Learning Street Talk (MLST)
Exploring Category Theory in Deep Learning
This chapter examines the role of category theory in enhancing geometric deep learning and graph neural networks, addressing the limitations of traditional methods. It introduces an innovative lecture series by Epeta that connects category theory with deep learning principles, emphasizing the need for accessible resources for newcomers. The discussion also highlights groundbreaking advancements, such as Natural Graph Networks, that challenge conventional structures and promote greater flexibility in neural network design.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
