Human Learning Optimization with Self-tuning Random Learning Strategy
Xuelian Hu, Chu Wang, Bowen Huang, Ling Wang · 2021
The human learning optimization (HLO) algorithm generates new solutions and searches for optimal solutions through random learning operator (RLO), individual learning operator (ILO) and social learning operator (SLO), of which the RLO is important for the exploration and exploitation abilities of HLO. To achieve an ideal balance between the exploration and exploitation and improve the search efficiency of the algorithm, this paper proposes an improved human learning optimization algorithm with self-tuning random learning strategy (HLOSRL), which enables the algorithm to automatically adjust the probability of RLO according to the feedback information. The proposed HLOSRL is compared with previous HLO variants as well as other recent meta-heuristics. The experimental results show that the HLOSRL has a better global searching ability.