Effective Exploration & Exploitation of the Solution Space via Memetic Algorithms for the Circuit Partition Problem

Memetic Algorithms (MAs) are Evolutionary Algorithms (EAs) that apply some sort of local search to further improve the fitness of individuals in the population. This paper provides a forum for identifying and exploring the key issues that affect the design and application of Memetic Algorithms. Several approaches of integrating Evolutionary Computation models with local search techniques (i.e Memetic Algorithms) for efficiently solving underlying VLSI circuit partitioning problem were presented. A Constructive heuristic technique in the form of GRASP was utilized to inject the initial population with good initial solutions to diversify the search and exploit the solution space. Furthermore, the local search technique was able to enhance the convergence rate of the Evolutionary Algorithm by finely tuning the search on the immediate area of the landscape being considered. Future work involves adaptive techniques to fine-tune parameter of the Genetic Algorithm and Local Search when combined to form a Memetic Algorithm. Balancing exploration and exploitation is yet another issue that needs to be addressed more carefully.