MachineLearning

Converts a vector of numbers into probabilities that sum to 1.

Squeezes any number into a range between 0 and 1.

When production data diverges significantly from the training dataset

When the production data is significantly different from the training distribution.

a technique for adding class imbalance by creating synthetic examples

an unsupervised clustering algorithm

The average absolute difference between predictions and labels, less sensitive to outliers than RMSE.

A regression metric that penalises large errors.

a subspace where the dimension is one less than the ambient space

a technique for reducing the dimensionality of data