Feb 192025 An opponent striatal circuit for distributional reinforcement learning D1-and D2-expressing striatal neurons encode separate parts of a learned reward distribution, paralleling modern approaches in machine learning.
