EPHE 245 Lecture Notes - Lecture 4: Reinforcement Learning, Tegmentum, Human Brain
Document Summary
Class 1. 32 neural basis of reinforcement learning. Provide examples of part and whole practice (detail) Rewrite will not be the same questions (picks different questions) Put a rat i(cid:374) a (cid:271)ox, put a (cid:449)ire i(cid:374) rat"s (cid:271)rai(cid:374), (cid:449)he(cid:374) the rat pushes the le(cid:448)er, triggers release so(cid:373)e dopamine. Then watch what the rats do (keep hitting the lever) Ignore sexual desire, ignore food, sleep then wake up and hit the lever. Look at the same model of old and milner. Monkeys usually given juice as reward and teach them to do certain tasks. Instead of stimulating dopamine, he measures dopamine (brain activity) Exp #2 (wolfram: monkey pushes a button, at some point they give reward or no reward, at the top is correct trials (reward), at the bottom is incorrect trials (no reward) Increase in firing for rewards, decrease in firing for punishments.