A Distributional Code for Value in Dopamine-Based Reinforcement Learning¶
Dabney, W., Kurth-Nelson, Z., Uchida, N., Starkweather, C. K., Hassabis, D., Munos, R., & Botvinick, M. (2020). A Distributional Code for Value in Dopamine-Based Reinforcement Learning. Nature, 577, 671-675.
Cited by¶
1 citation across 1 artifact.
Each citation links to the sentence it supports in the citing article.
Primes¶
- Reinforcement
- From machine learning back to neuroscience, deep-RL successes have shaped recent accounts of basal-ganglia learning, the transfer running in both directions over the same structure.
This sourceShows deep-RL distributional value accounts shaping models of dopamine and basal-ganglia learning.
Supported in partVerified against the work's full text
Dabney et al. show an AI distributional-RL idea (improving deep RL agents) prompting a new account of dopamine coding, atop the older AI-to-neuroscience RPE link; basal ganglia are barely mentioned.
“The RPE theory of dopamine derives from work in the artificial intelligence (AI) field of reinforcement learning... RL has made substantial advances... suggesting new hypotheses concerning reward-based learning in the brain”
- From machine learning back to neuroscience, deep-RL successes have shaped recent accounts of basal-ganglia learning, the transfer running in both directions over the same structure.
Verification¶
Does it exist? Confirmed. This work's DOI resolves to a registered record, which fixes its identity. That is all it fixes.
Does it back the claim? Read against the text for 1 of 1 citation: 1 supported in part. Each verdict is shown under its citation below, with what in the work backs the sentence.
Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.
See how references were verified.
Registry ID ref:8070d87e8c5c · see in the full table