Underfitting

Underfitting occurs when a machine learning model is too simple to capture the underlying patterns in the data. As a result, it performs poorly on both the training data and new, unseen data because it hasn't learned enough from the information provided. Think of it like using a very basic rule to make predictions, which doesn't accommodate the complexity of the actual data. To improve, more sophisticated models or additional features can help the model better understand the true relationships, leading to more accurate and reliable results.