Reward-based training of recurrent neural networks for cognitive and value-based tasks

Reward-based training of recurrent neural networks for cognitive and value-based tasks

Trained neural network models, which exhibit features of neural activity recorded from behaving animals, may provide insights into the circuit mechanisms of cognitive functions through systematic analysis of network activity and connectivity. However, in contrast to the graded error signals commonly...

Cur síos iomlán

Sonraí bibleagrafaíochta
Príomhchruthaitheoirí: H Francis Song, Guangyu R Yang, Xiao-Jing Wang
Formáid: Alt
Teanga:English
Foilsithe / Cruthaithe: eLife Sciences Publications Ltd 2017-01-01
Sraith:eLife
Ábhair:
recurrent neural network
reinforcement learning
prefrontal cortex
decision making
orbitofrontal cortex
working memory
Rochtain ar líne:https://elifesciences.org/articles/21492

Ar líne

https://elifesciences.org/articles/21492

Míreanna comhchosúla