Graph Neural Networks in Action: From Origins to the Age of Transformers

In this talk, Dr. Keita Broadwater, author of Graph Neural Networks in Action, traces the journey of machine learning with graphs—from early heuristic approaches to today’s powerful GNN architectures. We’ll explore why graph-structured data is uniquely valuable, how GNNs emerged as a breakthrough for modeling complex relationships, and where they stand in the current landscape dominated by transformer-based models.

As large language models and foundation models reshape the field, we’ll discuss the evolving role of GNNs and how they complement—or compete with—transformers in tasks like recommendation, drug discovery, and fraud detection. This session offers both historical context and forward-looking insights for anyone interested in the future of machine learning on structured data.