A Bayesian Neural Network-Based Method to Calibrate Microscopic Traffic Simulators

Table 1

Common notations.

Notation	Meaning
	The number of parameters for calibration
	The number of observed links
	The population size of the whole population in the th iteration
	The current fitness of the th individual in the th iteration
	The best fitness value (MSE) for all individuals in the population
	The maximum number of iterations
	The parameter group of the th individual in the th iteration including parameters
	The set of field-measured data
	The parameter group associated with
	The BNN outputs of the th individual in the th iteration