WiMi Hologram Cloud Inc, a leading global Hologram Augmented Reality (“AR”) Technology provider, announced that adaptive learning backtracking search algorithm (ALBSA) is proposed, which aims to improve the overall performance of BSA by introducing adaptive control parameters and novel mutation strategies. The adaptive control parameter adjusts the individual search step size based on the global and local information of the cluster in the current iteration, thus balancing the exploration and exploitation capabilities. The novel mutation strategy is based on different information guidance, which guides the mutation operation based on the global and local information of the swarm to improve the optimization capability of the algorithm. In addition, the introduction of multiple swarm strategies further enhances the adaptability of the algorithm to different search regions and the search capability.
WiMi improved the BSA through ALBSA to make it more competitive in optimizing problems. The goal of ALBSA is to improve the search efficiency and the quality of the solution through the introduction of adaptivity and flexibility, and the full use of global and local information. Experiments have demonstrated that ALBSA has better performance performance relative to traditional BSA and other evolutionary algorithms, allowing for better application and usefulness in real-world problems.
ALBSA proposed by WiMi has the following advantages over traditional BSA:
Adaptability and flexibility: An adaptive control parameter based on the global and local information of the swarms in the current iteration is designed to adjust the search step length of individuals. This allows the algorithm to better balance exploration and exploitation capabilities and adapt to the search needs of different problems.
Also Read : Cloud4C is Now a Google Cloud Specialized Partner for Infrastructure Services in APAC
Optimization capability enhancement: ALBSA introduces a novel mutation strategy based on the guidance of different information, which guides the mutation operation by making full use of the global and local information of the swarm. This enhances the optimization ability of the algorithm, allowing it to search the potential solution space in a more targeted way, improving the quality of the solution and the convergence speed.
Multiple swarm strategy: ALBSA implements a multiple swarm strategy, which means that multiple swarms are maintained at the same time, and each swarm can independently search a specific search region. This enhances the algorithm’s ability to search different search regions, thus exploring the solution space more comprehensively and increasing the probability of finding a globally optimal solution.
Competitiveness and effectiveness: Experimental results show that ALBSA is competitive and effective in solving optimization problems compared to traditional BSA and other optimization algorithms. It is able to find solutions with better performance higher convergence speed and search efficiency.
Overall, ALBSA improves on the traditional BSA by improving the adaptivity, optimization ability and searchability of the algorithm through the introduction of adaptive control parameters, novel mutation strategies and multiple swarm strategies. These advantages make ALBSA more application potential and effective in solving various optimization problems.
WiMi’s ALBSA, aims to optimize the BSA. ALBSA optimizes the search step size and optimization capability by designing adaptive control parameters and novel mutation strategies, and implements multiple swarm strategies to enhance the search capability for different search regions. The technical framework of ALBSA is as follows:
Initialization: Initialize the population and individuals, and set the initial values of the control parameters. Set other parameters of the algorithm, such as a maximum number of iterations, population size, etc.
Swarm information update: In each iteration, global information about the swarm is calculated based on the fitness of the current population, e.g., average fitness and optimal fitness. Local information of the swarm is computed based on the fitness of individuals, e.g., the relative fitness between an individual and its neighbors.
Adaptive tuning of control parameters: Adaptively adjusting control parameters based on global and local information about the swarm in the current iteration. Adjusting the control parameters can change the individual search step size to balance the exploration and exploitation capabilities of the algorithm.
New mutation strategy: Design a novel mutation strategy based on different information guidance to improve the optimization ability of the algorithm. The novel mutation strategy can determine the direction and magnitude of mutation based on the global and local information of the swarm in the current iteration.
Update and evaluation of solutions: Use control parameters and mutation strategies to update individual solutions. The updated solutions are evaluated and fitness values are calculated.
Termination condition check: Check whether the termination condition is satisfied, such as reaching the maximum number of iterations or finding a satisfactory solution. If the termination condition is satisfied, the algorithm ends; otherwise, return to step 2 for the next round of iteration.
ALBSA continuously adjusts the control parameters and mutation strategies to utilize global and local information to guide the search in order to improve the overall performance of the algorithm. Multiple swarm strategies further enhance the algorithm’s ability to search different search regions. Through experimental verification, ALBSA is competitive and effective compared with other evolutionary algorithms to find better solutions.
SOURCE : PRNewswire