Skip to main content

Table 1 Simulation study hyperparameters

From: Using a novel genetic algorithm to assess peer influence on willingness to use pre-exposure prophylaxis in networks of Black men who have sex with men

Hyperparameter

Value

Description

chromosomes

21

Number of chromosomes

probb

0.01

Initial probability of blending (\(p_b\))

factorb

2

Multiplicative factor for modifying \(p_b\)

maxb

0.2

Maximum value of \(p_b\)

iterb

1000

Number of iterations with no improvement before modifying \(p_b\)

probc

0.2

Initial probability of crossover (\(p_c\))

factorc

0.5

Multiplicative factor for modifying \(p_c\)

minc

0

Minimum value of \(p_c\)

iterc

1000

Number of iterations with no improvement before modifying \(p_c\)

probm

0.2

Initial probability of blending (\(p_m\))

factorm

0.5

Multiplicative factor for modifying \(p_m\)

minm

0.01

Minimum value of \(p_m\)

iterm

1000

Number of iterations with no improvement before modifying \(p_m\)

sigma

1

Initial value of standard deviation \(\sigma\) of error for mutation operator

factors

0.5

Multiplicative factor for modifying \(\sigma\)

mins

0.001

Minimum value of \(\sigma\)

iters

2000

Number of iterations with no improvement before modifying \(\sigma\)

max_iter

1E6

Maximum number of iterations to run algorithm

min_improve

0

Minimum decrease in value of objective function considered an improvement

min_dev

0

Acceptable value of objective function for stopping algorithm

reintroduce

“elite”

type of chromosome to be reintroduced

iterr

2500

Number of iterations with no improvement before reintroducing chromosome

  1. The table gives the name of each hyperparameter in the software developed to implement the new genetic algorithm, the value used for the simulation study and data analysis in the paper, and a description of the hyperparameter