Optimization of Neural Network Architecture using Lions Optimization Algorithm
Kaptan Malik, Kanishk Srivastava, Krishna Sharma · 2020
For several years, humans have observed nature to find patterns and similarities between their problems and natural occurrences. Due to this, there has been a surge of research and algorithms regarding our environment. It is usually a debilitating assignment to structure and manufacture an ANN for a selected issue such as nature. There is a pair of thorough venture rules that are accessible to characterize the ANN and, in any case, numerous parameters to switch. The productive techniques are an option in contrast to counterfeit neural systems planning; however, the computational cost required for specific applications is restrictive. Accordingly, the critical point here is to contemplate progressed developmental calculation strategies, which likewise advance accomplishing ANN structures that commit to the task efficiently, explicit to each issue using Lion's Optimization Algorithm (LOA). Pima Indians Diabetes Database used as a model for the optimization of ANN. The results show that subsequent generations of modeled lions provide better accuracy and efficiency to tend towards the highly optimized solution.