Prefrontal cortex as a meta-reinforcement learning system
| Autor: | Dhruva Tirumala, Dharshan Kumaran, Matthew Botvinick, Zeb Kurth-Nelson, Demis Hassabis, Hubert Soyer, Jane X. Wang, Joel Z. Leibo |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
0301 basic medicine
Dopamine Models Neurological Prefrontal Cortex Optogenetics 03 medical and health sciences 0302 clinical medicine Reward Artificial Intelligence medicine Canonical model Reinforcement learning Animals Humans Learning Computer Simulation Reinforcement Prefrontal cortex Cognitive science General Neuroscience Perspective (graphical) 030104 developmental biology Neuroscience research Psychology Neuroscience Reinforcement Psychology 030217 neurology & neurosurgery Algorithms medicine.drug |
| Zdroj: | Nature Neuroscience |
| ISSN: | 1546-1726 |
| Popis: | Over the past twenty years, neuroscience research on reward-based learning has converged on a canonical model, under which the neurotransmitter dopamine ‘stamps in’ associations between situations, actions and rewards by modulating the strength of synaptic connections between neurons. However, a growing number of recent findings have placed this standard model under strain. In the present work, we draw on recent advances in artificial intelligence to introduce a new theory of reward-based learning. Here, the dopamine system trains another part of the brain, the prefrontal cortex, to operate as its own free-standing learning system. This new perspective accommodates the findings that motivated the standard model, but also deals gracefully with a wider range of observations, providing a fresh foundation for future research. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink