Meta-Lamarckian learning in three stage optimal memetic exploration

Abstract

Abstract-Three Stage Optimal Memetic Exploration (3SOME) is a single-solution optimization algorithm where the coordinated action of three distinct operators progressively perturb the solution in order to progress towards the problem's optimum. In the fashion of Memetic Computing, 3SOME is designed as an organized structure where the three operators interact by means of a success/failure logic. This simple sequential structure is an initial example of Memetic Computing approach generated by means of a bottom-up logic. This paper compares the 3SOME structure with a popular adaptive technique for Memetic Algorithms, namely Meta-Lamarckian learning. The resulting algorithm, Meta-Lamarckian Three Stage Optimal Memetic Exploration (ML3SOME) is thus composed of the same three 3SOME operators but makes use a different coordination logic. Numerical results show that the adaptive technique is overall efficient also in this Memetic Computing context. However, while ML3SOME appears to be clearly better than 3SOME for low dimensionality values, its performance appears to suffer from the curse of dimensionality more than that of the original 3SOME structure.

Lamarckian learning‐based whale optimization algorithm to localize non‐line‐of‐sight nodes for achieving reliable data dissemination in vehicular ad‐hoc networks

“…In this phase of exploitation, the method of bubble‐net attacking strategy is employed for effectively determining the position of the NLOS nodes 27 . This method of bubble‐net attacking strategy incorporates the mechanism of shrinking encircling and spiral updating position as detailed below.…”

“…Thus, it may not represent the complete direction and level of evolution. Moreover, solving high dimensional optimization problem like the NLOS nodes' localization problem with multiple number of local optimum solutions is even more complex 27 . If the current best solution determined by a search agent is a local optimum solution, then the probability of the search agents to fall into the local optimum solution during the process of exploration is high.…”

“…The search agents start the process of iterative optimization from the individual solutions that are randomly initialized. 27 The convergence of the localization algorithm can be accelerated and its performance can be significantly improved to a considerable level, when the optimal solutions are closely situated with respect to some of the individual search agents. Further, this good point set theory has confirmed that the weighted sum of error related to m good points is comparatively less than that of the sum of estimated error determined for the any m other points during the application of approximate computation functions implemented in the r-dimensional Euclidean space unit cube.…”

“…Moreover, solving high dimensional optimization problem like the NLOS nodes' localization problem with multiple number of local optimum solutions is even more complex. 27 If the current best solution determined by a search agent is a local optimum solution, then the probability of the search agents to fall into the local optimum solution during the process of exploration is high. This limitation mainly occurs, since the position updating process of the remaining search agents completely concentrates on the current estimated local optimum point.…”

“…The search agents start the process of iterative optimization from the individual solutions that are randomly initialized 27 . The convergence of the localization algorithm can be accelerated and its performance can be significantly improved to a considerable level, when the optimal solutions are closely situated with respect to some of the individual search agents.…”