Divide and Conquer: A Quick Scheme for Symbolic Regression

Symbolic regression aims to find a function that best explains the relationship between independent variables and objective value based on given set of sample data. Genetic programming (GP) is usually considered as an appropriate method for problem since it can optimize functional structure coefficients simultaneously. However, convergence speed GP might be too slow large scale problems involve number variables. Fortunately, in many applications, target separable or partially separable. This feature motivated us develop new method, divide conquer (D&C), symbolic regression, which divided into sub-functions are then determined by any algorithm. The separability probed proposed technique, Bi-Correlation test (BiCT). D&C powered has been tested some real-world study shows help get much more rapidly.