A CIFAR workshop took place on Tuesday, the 18th, and Wednesday, the 19th in Tübingen, in conjunction with the ELLIS Institute Tübingen and the Max Planck Institute for Intelligent Systems (MPI-IS).
Hosted at MPI-IS, the workshop delved into the robust and causal learning of world models, with a particular focus on inspirations from neuroscience and applications in AI safety. Specifically, it aimed to:
- Gain an in-depth understanding of causal world models and their significance for understanding and advancing interactive AI systems;
- Explore the role of world models in neuroscience to better comprehend how the brain processes information and adapts to new situations; and,
- Discuss the challenges and considerations in building world models for AI safety.
Featuring award-winning scientists and accomplished specialists in AI, the workshop was a blend of intriguing talks and engaging panel discussions. Top scientists had the opportunity to present groundbreaking research and explore future collaborations.
Find the full agenda and list of speakers here.