What are some common hyperparameter tuning techniques used in deep learning?

Common hyperparameter tuning techniques include Grid Search, Random Search, and Bayesian Optimization. Grid Search systematically tests a predetermined set of hyperparameter values, ensuring comprehensive coverage of the search space but can be computationally expensive. Random Search samples hyperparameter combinations randomly, often yielding better results faster than Grid Search. Bayesian Optimization uses probabilistic models to identify promising hyperparameters more efficiently by focusing on regions of the space that have shown better performance, offering a more informed and effective tuning process.