Research Review for Broad Learning System Algorithms, Theory, and Applications
Abstract : In recent years, the appearance of the broad learning system (BLS) is poised to revolutionize conventional artificial intelligence methods. It represents a step toward building more efficient and effective machine-learning methods that can be extended to a broader range of necessary research fields. In this survey, we provide a comprehensive overview of the BLS in data mining and neural networks for the first time, focusing on summarizing various BLS methods from the aspects of its algorithms, theories, applications, and future open research questions. First, we introduce the basic pattern of BLS manifestation, the universal approximation capability, and essence from the theoretical perspective. Furthermore, we focus on BLS's various improvements based on the current state of the theoretical research, which further improves its flexibility, stability, and accuracy under general or specific conditions, including classification, regression, semisupervised, and unsupervised tasks. Due to its remarkable efficiency, impressive generalization performance, and easy extendibility, BLS has been applied in different domains. Next, we illustrate BLS's practical advances, such as computer vision, biomedical engineering, control, and natural language processing. Finally, the future open research problems and promising directions for BLSs are pointed out.
? A large number of experiments show that our algorithm can effectively solve the problem of transfer learning in image classification better than some existing algorithms.
? Many factors can prevent it from achieving the maximum, even if one exists. In other words, optimization is not the same a optimality.
? Reinforcement learning’s connection to optimization methods deserves some additional comment because it is a source of a common misunderstanding.
? A typical evolutionary method would hill-climb in policy space, successively generating and evaluating policies in an attempt to obtain incremental improvements.
? It is just as applicable when behavior continues indefinitely and when rewards of various magnitudes can be received at any time.
? It means that the adoption of advanced manufacturing technologies, such AI and ML, is an emerging issue.
? In other words, AI/ML algorithms represent an opportunity to handle high dimensional problems and data.
? Artificial intelligence can be useful to solve critical issue for sustainable manufacturing (e.g., optimization of energy resources, logistics, supply chain management, waste management, etc.)
? The perspective of sustainability, the analysis highlighted that the new paradigm of smart manufacturing has the potential to bring fundamental improvements in the industry by addressing the issue of scarce resources and improving productivity.
• The proposed algorithm can learn a robust classification model by using a small part of labeled data from the target domain and all labeled data from the source domain.
• The proposed algorithm inherits the computational efficiency and learning capability of BLS.
• In this paper, an algorithm called domain adaptation broad learning system based on locally linear embedding (DABLS-LLE) is proposed.
• Different from many deep neural networks which use gradient descent method to update weights, our proposed algorithm can use pseudo inverse to get the optimal solution directly.
• The proposed DABLS-LLE algorithm contains three stages: reconstruction weights calculating, BLS feature mapping, output weights learning.
? The indicators chosen to perform the analysis were total papers (TPs), which is the total number of publications, and total citations (TCs), which is the total number of citations.
? The analysis was done using the Web of Science and SCOPUS database. Furthermore, UCINET and NVivo 12 software were used to complete them.
? Its usage has spread to various fields, such as learning machines, which are currently used in smart manufacturing, medical science, pharmacology, agriculture, archeology, games, business, and so forth.
? It can be noted that the most used term is precisely “machine”, “learning”, and “intelligence”, which the software represents with greater characters than all the other terms.
|