Lance Da Costa, a PhD candidate at Imperial College London, dives into the fascinating intersection of cognitive systems and AI. He discusses his work on the free energy principle, arguing that all intelligent agents minimize free energy for perception and decision-making. The conversation covers the advantages of active inference over traditional AI methods, highlighting its potential for safety and explainability. Lance also examines the challenges of structured learning and the role of Bayesian model reduction in adapting agents to changing environments.

Mathematician Lance Da Costa delves into world and self models in active inference, stochastic differential equations, solenoidal flows in living systems, recognition density, meditation attention, prior beliefs, and cutting-edge vision technology in this insightful podcast.