Underfitting and Overfitting in ML

In Machine Learning when the model performs well on the training data but does not generalize well then we call it overfitting. Overfitting happens when the model is too complex relative to the amount and noisiness of the training data.

The possible solutions to overfitting are by:

• Simplifying the model by selecting one with fewer parameters (e.g., a linear model rather than a high-degree polynomial model), reducing the number of attributes in the training data, or by constraining the model
• Gathering more training data
• Reducing the noise in the training data (e.g., fix data errors and remove outliers)

You may say underfitting is the opposite of overfitting. Or we can say the condition as if we are trying to fit undersized clothes. Underfitting occurs when our machine learning model is not able to capture the underlying trend of the data. It occurs when your model is too simple to learn the underlying structure of the data.

The possible solutions for underfitting are by:

• Selecting a more powerful model, with more parameters
• Feeding better features to the learning algorithm (feature engineering)
• Reducing the constraints on the model (e.g., reducing the regularization hyperparameter)