concept dropout in category machine learning
appears as: dropout

This is an excerpt from Manning's book Machine Learning with TensorFlow, Second Edition.
Dropout—TensorFlow comes with a handy tf.nn.dropout function, which can be applied to any layer of the network to reduce overfitting. It turns off a randomly selected number of neurons in that layer during training so that the net- work must be redundant and robust to inferring output.
The application of a technique called dropout which randomly turns off neurons in a particular layer so that the architecture learns a more resilient representation of the input.