International Research Journal of Education and Technology

Hyperparameter optimization is a crucial aspect of machine learning that directly impacts model performance. Traditional hyperparameter tuning methods, such as grid search and random search, are computationally expensive and often fail to find optimal configurations efficiently. Automated Hyperparameter Optimization (HPO) using AI has emerged as an advanced solution to enhance model accuracy and reduce computational overhead.

This paper explores various AI-driven HPO techniques, including Bayesian optimization, reinforcement learning, evolutionary algorithms, and neural architecture search (NAS). Furthermore, we discuss the integration of HPO in Auto ML frameworks, its impact on model efficiency, and future research directions in hybrid optimization strategies. The adoption of AI-driven HPO techniques enables more efficient model selection, improving scalability and generalization across diverse machine learning applications.

Keywords – Hyperparameter Optimization, Auto ML, Bayesian Optimization, Neural Architecture Search, Reinforcement Learning.