Graphs and Transformers - Exploring the Next AI Frontiers

Explore the cutting-edge developments in artificial intelligence through this 45-minute conference talk from the Data Science Festival. Delve into the evolution of deep learning models, focusing on the emergence of graphs and transformers as fundamental building blocks in modern AI architectures. Discover how these innovations have revolutionized natural language understanding, protein research, and image generation. Learn about the spectacular progress in deep learning over the past 15 years, driven by massive computational power and leveraging key abstractions. Gain insights into the current state of AI technology and its potential future directions. Following the main presentation, engage with a panel of experts as they discuss and debate the future of geometric deep learning, transformers, and the role of immense computational power in shaping AI advancements.