R/multiStart.r

In neurodata/graphstats: Graph Statistics

#' Multiple Restart SGM
#'
#' Runs SGM algorithm with random starting point generated by \code{rsp} \code{R}
#' times and generates an average probability matrix for matched vertex
#' over the \code{R} restarts
#'
#' @param A adjacency matrix for \eqn{G_1}
#' @param B adjacency for correlated matrix \eqn{G_2}
#' @param R number of restarts
#' @param s number of seeds; assumed to be first \eqn{s} vertices in both
#' \eqn{G_1} and \eqn{G_2}
#' @param gamma (upper bound on alpha) should be in (0,1), max tolerance
#' for alpha, how far away from the barycenter user is willing to go for
#' the initialization of \code{sgm} on any given iteration
#' @param pad a scalar value for padding for sgm (cases where two graphs
#' have different number of vertices); defaulted to 0
#' @param iter number of iterations allowed to be used in sgm step
#'
#' @return \eqn{n \times n} average probability matrix over all restarts.
#'
#' @author Heather Gaddy Patsolic <hgaddy1@jhu.edu>
#' @export

multiStart <- function(A,B,R,s,gamma,pad=0,iter=20){
  n <- max(nrow(A),nrow(B))
  P <- matrix(0,n,n)

  for(i in 1:R){
    M <- rsp(n-s,gamma)
    Ps <- sgm.ordered(A,B,s,start=M,pad=pad,maxiter=iter)
    P <- P + Ps
  }

  D <- (1/R)*P
  return(D)
}