An implementation of adaptive probabilistic mapping is actor. An actor interacts with an application program or environment by means of exchanging signals. Specific arguments of an adaptive probabilistic mapping correspond to input signals of an actor. Specific outcomes of an adaptive probabilistic mapping correspond to output signals of an actor. Input signals of an actor can be its transformed output signals similarly to the case when the arguments of an adaptive probabilistic mapping are its transformed outcomes.
An actor usually receives input signals, spur (see Spur-Driven Behavior) increments, and time increments and emits output signals with the goal to maximize spur increment velocity. To achieve that goal, the actor performs basic forecasting of spur increments resulting from emitting various output signals. The author believes that by combining multiple actors or applying other approaches it is possible to amplify this forecasting capability to create a system intelligently interacting with an environment of real-world complexity.