Multimodal, Federated, Agentic, and Lifelong Learning for Autonomous Machine Intelligence

The proposed workshop "Multimodal, Federated, Agentic, and Lifelong Learning for Autonomous Machine Intelligence" aims to explore emerging foundations, models, and systems for building truly autonomous intelligent machines capable of learning, reasoning, and acting over long time horizons with minimal human supervision.

Recent advances in self-supervised learning, world models have opened new directions toward autonomous intelligence. At the same time, the rise of agentic Al, multi-agent systems, large language and vision models, and federated learning introduces new challenges in representation learning, coordination, memory, and continual adaptation.