credit assignment
The machine learning problem of determining which actions contributed to an outcome.
1 article
Last mentioned Credit assignment is the problem of determining which of many actions or computations contributed to the success or failure of a final outcome. It is especially important in reinforcement learning, where rewards arrive only after a sequence of actions, and it also refers to deciding how much each weight in a neural network is responsible for an error.
This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.