Adaptive Randomness: A New Population Initialization Method

Abstract EN

Population initialization is a crucial task in population-based optimization methods,which can affect the convergence speed and also the quality of the final solutions.

Generally,if no a priori information about the solutions is available, the initial populationis often selected randomly using random numbers.

This paper presents a new initializationmethod by applying the concept of adaptive randomness (AR) to distribute theindividuals as spaced out as possible over the search space.

To verify the performanceof AR, a comprehensive set of 34 benchmark functions with a wide range of dimensionsis utilized.

Conducted experiments demonstrate that AR-based population initializationperforms better than other population initialization methods such as random populationinitialization, opposition-based population initialization, and generalized opposition-basedpopulation initialization in the convergence speed and the quality of the finalsolutions.

Further, the influences of the problem dimensionality, the new controlparameter, and the number of trial individuals are also investigated.