lags and rolling window in Azure AutoML Time series forecasting

I'm following this tutorial to try Machine Learning AutoML Forecasting.

In the several parameters we can submit to the AutoML experiment, we have these ones:

1. target_logs;
2. target_rolling_window_size;

Can you explain with an example how the several forecasting algorithms works when these two parameters are set?

Thank you

automl_advanced_settings = {
    'time_column_name': time_column_name,
    'max_horizon': max_horizon,
    'target_lags': 12,
    'target_rolling_window_size': 4,
}
automl_config = AutoMLConfig(task='forecasting',                             
                             primary_metric='normalized_root_mean_squared_error',                  

                             experiment_timeout_hours=0.3,
                             training_data=train,
                             label_column_name=target_column_name,
                             compute_target=compute_target,
                             enable_early_stopping = True,
                             n_cross_validations=3,                             
                             verbosity=logging.INFO,
                            **automl_advanced_settings)