Deriving Controllable Local Optimal Solutions through an Environment Parameter Fixed Algorithm

This paper addresses the challenge of optimizing objective functions in engineering problems influenced by multiple environmental factors, such as temperature and humidity. Traditional modeling approaches often struggle to capture complexities non-ideal situations. In this research, we propose a novel approach called Environment Parameter Fixed Algorithm (EPFA) for function deep neural network (DNN) trained specific environment. By fixing parameters DNN defined function, transform original optimization problem into control parameter problem. We integrate EPFA-CLS (Controllable local-Optimal Solution) with Gradient Descent algorithms Adagrad obtain optimal solution. To demonstrate concept, apply our an course model validate it using Boston house price datasets. The results effectiveness handling complex environments, offering promising outcomes practical applications.