Reinforcement Learning in the Brain 🧠 | Harvard University Lecture | Spring 2022
A lecture on dopamine and the temporal difference (TD) learning algorithm for GenEd1125: Artificial and Natural Intelligence at Harvard University
TIMESTAMPS
0:00 Intro
1:38 Dopamine
5:55 A fundamental puzzle
9:05 Dopamine signals a reward prediction error
15:58 Temporal difference (TD) learning
21:56 Example application of TD algorithm to Schultz et al. 1997
37:49 TD learning in other environments
39:42 Takeaways on TD learning
44:38 How does TD learning fit into the rest of the brain's learning circuitry?
46:30 Actor-critic models
49:40 When reward learning goes awry - Cocaine addiction in Redish 2004
tags: #neuroscience #harvard #brain #lecture #students #ai #reinforcementlearning #free #education #educational #educationalvideos #moocs #lucylai