actuator (macula_tweann v0.18.1)

Actuator process for TWEANN networks.

Actuators are the output interface of a neural network. They receive signals from neurons and produce actions or outputs that affect the environment or are used for fitness evaluation.

Actuator Lifecycle

1. Spawned by cortex with configuration 2. Waits for signals from input neurons 3. Accumulates all inputs 4. Calls actuator function to produce output 5. Reports output to cortex 6. Repeats from step 2

Actuator Functions

Actuator functions determine how the neural output is used. Common examples:

- pts - Pass-through sum (for testing) - xor_output - XOR problem output - pole_output - Pole balancing action

Summary

Functions

init(Opts)

Initialize the actuator and enter the main loop.

start_link(Opts)

Start an actuator process.

Functions

init(Opts)

-spec init(map()) -> no_return().

Initialize the actuator and enter the main loop.

start_link(Opts)

-spec start_link(map()) -> {ok, pid()}.

Start an actuator process.

Options: - id - Unique identifier for this actuator - cortex_pid - PID of the controlling cortex - actuator_name - Atom naming the actuator function - vector_length - Expected length of input vector - fanin_pids - List of PIDs that send input to this actuator - scape_pid - PID of the scape/environment (optional) - parameters - Additional parameters for actuator function