
DR. JEFF BECK - THE BAYESIAN BRAIN
Machine Learning Street Talk (MLST)
The Components and Challenges of Active Inference
Active inference consists of an inference engine, a prediction model, and a reward function to motivate behavior. The prediction model makes predictions about the world based on past data, while the inference engine learns and identifies that model. However, the source of the reward function remains a challenging question. Active inference suggests that motivating behavior should be based on maintaining homeostatic equilibrium and the statistics of the blanket, rather than relying on a predefined reward function.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.