|
Algorithm in a jar? Picking apart the computations behind learning and decision making| old_uid | 12271 |
|---|
| title | Algorithm in a jar? Picking apart the computations behind learning and decision making |
|---|
| start_date | 2013/03/29 |
|---|
| schedule | 12h30-13h30 |
|---|
| online | no |
|---|
| summary | Reinforcement learning (RL) is one of the most successful theories in neuroscience. It has been used to explain myriad behavioural findings and its neural signature - in the form of a prediction error signal - has been found in the firing of midbrain dopaminergic neurons. Despite this success, many fundamental questions remain about the nature of RL in the brain. In this talk I will focus on two of these questions: 1) How do we adapt our learning in the face of change? and 2) How do we adapt our decision making to facilitate learning? Answers to these questions suggest extensions to RL theory and suggest a computational role for norepinephrine in modulating behaviour. |
|---|
| responsibles | Koechlin |
|---|
| |
|