Q-learning Machine Learning
noun phrase
Definition: A value-based reinforcement learning algorithm that learns the optimal action-value function by updating Q-values according to the Bellman optimality equation [Google ML Glossary].
Example in context: “Q-learning for instance is based on TD-errors and the Bellman equation.” [Rajan et al. 2023]
Related terms: tabular Q-learning; reinforcement learning; Q-function; Bellman equation