Green Machine Learning (GML) focuses on optimizing machine learning processes to further improve energy efficiency and sustainability without affecting its performance. This approach integrates various studies that point toward striking a balance between accuracy, energy consumption, and carbon footprint. Key strategies include using adaptive frameworks for hyperparameter optimization that seek to optimize energy efficiency, such as FanG-HPO and MISO-AGP. Besides, reinforcement learning is used for energy-aware indoor localization, bringing out the essential aspect of minimizing energy costs for practical applications. Therefore, evaluation metrics related to energy consumption, carbon emission, and model performance (accuracy, precision, recall) are important for any GML approaches. GML intends to yield insights into sustainable AI practices by incorporating a variety of datasets and using the latest optimization techniques, hence guiding scalable and greener machine learning solutions.