RETRACTED ARTICLE: Deep reinforcement learning for comprehensive route optimization in elastic optical networks using generative strategies

The latest advances in Deeper Reinforcement Learning (DRL) have completely changed how decision-making and automatic control issues are solved. The study community increasingly applies DRL methods to networking-related optimization issues like routing. Previous suggestions, though, frequently came short of conventional routing methods and could not produce satisfactory outcomes. Because of the constant development of one network efficiency parameter at the cost of individuals, most conventional safeguarding and restoring techniques will become ineffective. We believe that collectively considering the primary network parameters will be more advantageous for thorough network efficiency optimization. Additionally, elastic optical networking (EONS)’ highly adaptive characteristics necessitate the development of innovative machine learning-driven systems that adjust to the constantly changing nature of operations to execute the analysis. This study investigates how to develop DRL agents for resolving a route optimization issue using a generative strategy (GS). Our research findings indicate DRL agents operate better when employing our unique description.