WiMi Developed Deep-Mining Backtracking Search Optimization Algorithm Guided by Collective Intelligence

WiMi Hologram Cloud, a leading global Hologram Augmented Reality (“AR”) Technology provider, announced that a collective intelligence guided backtracking search optimization algorithm for deep mining(CIGBSA) based on the backtracking search optimization algorithm (BSA) is developed aiming to solve the problem of insufficient development capability of traditional BSA.

AiThority Insights: From Earth to the Stars: Unleashing the Power of AI in Space Engineering

The core of CIGBSA is deep mining guided by collective intelligence. Based on the retrospective search optimization algorithm, the optimal tendency of collective intelligence is introduced to accelerate the search process and improve the performance of the algorithm by tapping into the advantages of collective intelligence. To guide CIGBSA’s decision-making in searching for the vertices of the best individuals, CIGBSA designs learning operators based on topological oppositions. These operators guide the algorithm in searching and learning based on the topology of the current solution space. By analyzing and modeling the topology, CIGBSA is able to more accurately assess the quality of the solution and evolve towards a globally optimal solution.

The linear combination strategy of CIGBSA plays an important role in guiding individuals towards the optimal solution by introducing difference vectors. By analyzing the differences and similarities between individuals, the developers of WiMi have designed an effective linear combination strategy that enables individuals to learn and evolve more efficiently. The use of this strategy accelerates the convergence speed of the algorithm and improves the quality of the solution. In order to balance the overall performance, CIGBSA simulates a clustering-trending strategy with collective intelligence. In the linear combination strategy, WiMi developed another difference vector that guides individuals to learn from the average of the current generation. The introduction of this strategy allows individuals to take into account both global and local information in the learning process, thus improving the robustness and adaptability of the algorithm.

WiMi has carried out extensive demonstrations in validating the performance of CIGBSA, choosing a series of standard optimization problems and real-world application problems as test benchmarks and comparing them with the original BSA as well as state-of-the-art algorithms. The performance and competitiveness of CIGBSA are evaluated by comparing the convergence speed of the algorithm, the quality of the solution and the robustness of the algorithm. During the experimental process, repeated parameter tuning and optimization are carried out to further enhance the performance of CIGBSA, and the optimal combination of parameters is searched by adjusting parameters such as the learning rate, the weights of the difference vectors, and the topologically-opposed learning operators, so that the algorithm can achieve optimal results on different problems.

Read More about AiThority: Money Meets Machine: 5 Game-Changing AI Trends in Finance and Insurance

WiMi’s CIGBSA has several advantages over the traditional BSA:

Simple structure and easy to implement: The structure of CIGBSA is relatively simple and easy to understand and implement. It is based on the BSA and introduces the ideas of ensemble intelligence and deep mining, which makes the algorithm have global exploration ability and efficient learning ability.

Strong global exploration ability: CIGBSA is able to explore the solution space more comprehensively through deep mining guided by ensemble intelligence, thus increasing the probability of discovering a globally optimal solution. Compared with traditional BSA, CIGBSA can converge to the neighborhood of the optimal solution more quickly when solving optimization problems.

Improved Learning Mechanisms: CIGBSA introduces topological dyadic learning operators and linear combination strategies that enable the algorithm to more accurately assess the quality of the solution and guide the learning of individuals. This learning mechanism accelerates the learning process of the algorithm, allowing individuals to adapt faster to the characteristics of the problem, improving the quality of the solution and the performance of the algorithm.

Good convergence and robustness: CIGBSA realizes the ability of an individual to take into account both global and local information in the learning process through a clustering-trending strategy that simulates collective intelligence. The introduction of this strategy gives the algorithm good convergence and robustness, and enables it to find high-quality solutions stably under different problems and parameter settings.

Competitiveness and wide application: CIGBSA is experimentally competitive with BSA and state-of-the-art algorithms. It shows excellent performance on several standard optimization problems and practical application problems. Therefore, CIGBSA is expected to play an important role in the fields of engineering optimization, resource allocation, and data analysis, and to provide efficient and accurate solutions for solving practical problems.

To summarize, WiMi’s CIGBSA has advantages in terms of structural simplicity and global exploration capability, efficient learning and evolution through improved learning mechanisms and strategies, as well as good convergence and robustness. Its competitiveness and potential for a wide range of applications make it a powerful tool for solving optimization problems. The development of this algorithm opens up new avenues for solving optimization problems and is expected to have a wide impact in practice and provide efficient and accurate solutions for various industries.

 Latest AiThority Insights : From Spam-Free to Superb: Google’s Take on AI-Generated Content

 [To share your insights with us, please write to sghosh@martechseries.com]