Improving evolutionary algorithm performance on maximizing functional test coverage of ASICs using adaptation of the fitness criteria

Abstract

Adaptation of the fitness criteria can be a very powerful tool, enhancing the feedback scheme employed in standard evolutionary algorithms. When the problem the evolutionary algorithm (EA) is trying to solve is changing over time, the fitness criteria need to change to adapt to the new problem. Significant performance improvements are possible with feedback based adaptation schemes. This work outlines the results of an adaptation scheme applied to maximization of the functional test coverage problem.