Application of recent optimization algorithms to the cancer risk assessment problem.

In this paper, the use of heuristic algorithms for solving the problem of maximum likelihood estimation of parameters, when data is censored, is explored. We apply modern heuristics that have successfully been used to solve difficult combinatory problems. In particular we use the genetic algorithms and evolutionary programming to compare with the simplex procedure. The data was obtained by simulation. Finally, we discuss the results obtained and the advantage of applying the heuristics to the maximum likelihood problem.