Memetic Estimation of Distribution Algorithm (MEDA)

Finding an efficient solution for solving the optimization problems has always been among the major challenges in the field of artificial intelligence. A number of algorithms such as the Genetic Algorithm and the Estimation of Distribution Algorithm (EDA) have been proposed for the purpose of solving such problems. The current study identifies the strengths and weaknesses of the EDA and describes a more complete form, called the MEDA, as an enhanced form of the EDA. MEDA is a more complete method and like the EDA, makes use of a probability vector. The main advantage of the Estimation of Distribution Algorithm is maintaining the building blocks. However, to the extent that it has advantages in that area, it has weaknesses in the local search and does not show a considerable strength in that area. In order to overcome this problem, the proposed method make use of the SLS method. Based on the results obtained through implementing the proposed MEDA method on the PSP problem, it was found that the proposed method shows enhanced efficiency and provides more accurate results.

Keywords:

Estimation of Distribution Algorithm (EDA) , Protein Structure Prediction (PSP) , Stochastic Local Search (SLS) , The probability model.