Workshop: Learning Dynamics in Natural and Artificial Intelligence: Evolution, Adaptation, and the Foundations of Efficient Learning

This workshop will convene researchers from artificial intelligence, neuroscience, cognitive science, and related disciplines to examine the principles governing learning and training dynamics across natural and artificial systems. 

Organized around themes at the intersection of evolution, biological intelligence, and AI, the meeting will explore how biological and artificial systems each achieve — or fail to achieve — efficiency, flexibility, robustness, and generalization.  Researchers will explore how inductive bias, adaptation, inheritance, developmental constraints, selection, and multi-agent interaction may help illuminate the mechanisms underlying these similarities and differences and consider the forces that drive systems toward more generalized, versatile intelligence versus narrower specialization. 

By bringing these perspectives into dialogue, the workshop aims to clarify shared theoretical questions, identify productive points of methodological and conceptual overlap, and stimulate new interdisciplinary collaborations on the foundations of learning in both biological and artificial domains.

Thematic areas: 

1. Comparative learning in biological and artificial systems
   Why humans and other biological systems learn with greater efficiency, flexibility, and robustness than current machine learning models. 
2. Training dynamics and inductive bias
   How architectural constraints, priors, and learning dynamics shape what systems can learn, how quickly they learn, and how well they generalize. 
3. Evolution, inheritance, and selection
   What evolutionary processes can teach us about the emergence of adaptive behavior, structure, and learning efficiency in both natural and artificial systems. 
4. Multi-agent learning and adaptation
   How interaction among agents, environments, and populations influences learning, coordination, competition, and the development of complex behavior. 
5. Foundations for next-generation learning systems
   How cross-disciplinary perspectives might inform the design of more efficient, adaptive, and robust artificial intelligence systems.