Q-table
A Q-table is a lookup table storing estimated action values for state-action pairs in a tabular reinforcement-learning system.
Grokking AI Algorithms, Second Edition
A Q-table is a tabular reinforcement-learning model that stores estimated values for actions in particular states.
- How it is used
- Guides action selection in a Q-learning navigation simulation by scoring available moves for the current state.
- How deep it goes
- Explains rows, columns, action selection, initialization, training, and testing, without developing a mathematical derivation.
- Examples
- Uses a parking-lot navigation simulation, diagrams of states and action values, and a Python training function.
- What you can do after
- Apply it via a Q-learning simulation to train action scores and choose moves from a trained table.
Read how Grokking AI Algorithms, Second Edition explains Q-table →
Evolutionary Deep Learning: Genetic algorithms and neural networks
A Q-table is a tabular representation that stores estimated action-quality values for states in a Q-learning agent.
- How it is used
- Stores action-quality estimates that guide an agent’s next action in discrete reinforcement-learning environments.
- How deep it goes
- Defines the table layout, initialization, action policy, update equation, training loop, and exploration schedule without deriving Q-learning theory.
- Examples
- Uses OpenAI Gym’s FrozenLake environment, NumPy arrays, epsilon-greedy selection, training episodes, and evaluation runs.
- What you can do after
- Builds a tabular Q-learning agent, train it across episodes, and inspect its learned action values.
Read how Evolutionary Deep Learning: Genetic algorithms and neural networks explains Q-table →
Related concepts
- language modelkind of · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
- Q-learningused by · Q-learning is a model-free reinforcement-learning algorithm that estimates state-action values through updates based on rewards and subsequent state values.
- action-value functionhas part · An action-value function maps a state-action pair to the expected cumulative return from taking that action and subsequently following a specified policy.
- reward signalmeasures · A reward signal is an evaluative feedback value representing the desirability or payoff of an action, outcome, or state.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.