concept countermeasure in category deep learning
appears as: countermeasures, countermeasures, countermeasure

This is an excerpt from Manning's book Deep Learning with JavaScript: Neural networks in TensorFlow.js.
During the training of a machine-learning model, we want to monitor how well our model is capturing the patterns in the training data. A model that doesn’t capture the patterns very well is said to be underfit; a model that captures the patterns too well, to the extent that what it learns generalizes poorly to new data, is said to be overfit. An overfit model can be brought back on track through countermeasures such as regularization. In this section, we’ll show how visualization can help us spot these model behaviors and the effects of the countermeasures.