Extreme Learning Machine: A Comprehensive Survey of Theories & Algorithms

The Extreme Learning Machine (ELM), a quick and effective approach for training single-hidden-layer feedforward neural networks, is thoroughly reviewed in this paper. The objective is to provide a systematic review and analysis of the various theories, algorithms, and applications of ELM, which have not been adequately covered in the literature. The study examines the theoretical foundations of ELM, provides an overview of various ELM algorithms, analyzes the applications of ELM in different domains, and identifies future research directions and challenges in the field of ELM. The findings suggest that ELM is a promising and effective learning algorithm with many benefits as improved learning speed, good generalisation performance, and low complexity. The implications of this research for practice and research are significant. However, more empirical studies are necessary to evaluate its performance in real-world applications, and further research is required to assess its effectiveness in different domains. Future work should focus on exploring the potential of ELM in deep learning, enhancing the interpretability of ELM models, and developing new algorithms that can improve the learning efficiency of ELM. Overall, this survey provides a valuable reference for researchers and practitioners to better understand the potential and limitations of the ELM algorithm.