Selective network discovery via deep reinforcement learning on embedded spaces

Applied Network Science

Table 2 Detailed list of parameter values used for synthetic networks

Model	Type	Parameters
SBM	Background	\(k=[1,10], p_i=[0.01,0.4],r=[0.005,0.25],i=1 \ldots k\)
LFR	Background	\(\tau _1=[3,2], \tau _2=(1,1.9], \mu =[0.1,0.4], \langle d \rangle =[32,256]\), \(d_{\max }=[256,2048],\min _{c}=[256,1000], \max _{c}=[512,2000]\)
ER	Foreground	\(n_f=\{30,40,80\},k_f=\{1,2,4\},p_f=[0.5,1]\)

Number of nodes is represented by \(N=4000\). SBM parameters are: k represents the number of communities, \(p_{i}\) the edge probability for within-community i, r the across-community edge probability, such that \(p_{i} > r\). LFR parameters are: \(\tau _1,\tau _2\) skewness parameters for degree and cluster size distributions respectively, \(\langle d \rangle\) represents the average network degree, \(d_{\min },d_{\max }\) represent the min and max values of degree distribution, \(\min _c\) and \(\max _c\) represent the sizes of smallest and largest clusters, and finally \(n_f,k_f,p_f\) represent the size of the foreground subnetwork, number of foreground subnetworks and its edge probability, respectively