Results for "action mapping"
Behavior Cloning
Advanced
Learning action mapping directly from demonstrations.
Model
Intermediate
A parameterized mapping from inputs to outputs; includes architecture + learned parameters.
Policy
Intermediate
Strategy mapping states to actions.
SLAM
Intermediate
Simultaneous Localization and Mapping for robotics.
Value Function
Intermediate
Expected cumulative reward from a state or state-action pair.
Q-Function
Intermediate
Expected return of taking action in a state.
Agent Loop
Intermediate
Continuous cycle of observation, reasoning, action, and feedback.
Dynamics Model
Advanced
Predicts next state given current state and action.
Trajectory Optimization
Advanced
Optimizing continuous action sequences.
Action Space
Intermediate
Set of all actions available to the agent.