A Novel Conjugate Gradient Algorithm for Unconstrained Optimization and Its Application in COVID-19 Data Parameterization

To solve unconstrained optimization problems, this study proposes a conjugate gradient (CG) algorithm that satisfies both the convergence and descent conditions. The technique improves on conventional CG techniques through guaranteeing quick convergence and increased solution accuracy across a range of test functions. The study also applies this CG method to model COVID-19 transmission dynamics within a parameterized optimization scope. Analyzing reported cases from January 28, 2024, to September 29, 2024, demonstrates the model's effectiveness in capturing nonlinear trends, with a resurgence of COVID-19 cases in the latter half of the study period. This emphasizes the need for adaptive public health strategies in response to fluctuating infection rates, highlighting the significance of advanced mathematical modeling in infectious disease management.