# Temporal-Difference Learning **Domain:** Machine Learning / Reinforcement Learning **Doc Type:** Concept **Maturity:** Developed **Related:** [[Reward Prediction Error]], [[Richard Sutton]], [[Andrew Barto]], [[Reinforcement Learning]], [[Schedule and Loop]] ## Definition Temporal-difference learning updates value estimates from the difference between successive predictions before a final outcome is known. ## Schedule–Loop Context It joins behaviorist reinforcement vocabulary to cybernetic error correction, allowing a model to learn continuously from discrepancies between expectation and experience. ## Backlinks - [[Schedule and Loop]] - [[articles/Schedule and Loop|Schedule and Loop]] - [[Russia and Prussia Kybernetiks]]