Wrapper methods are a feature selection technique in machine learning that evaluate the performance of different subsets of features to identify the optimal combination for a given model. They are computationally intensive but often yield better results as they consider the interaction between features and the model's learning algorithm.