 What is the curse of dimensionality in very high dimensional space in machine learning in many other fields, the relative distances between points almost vanishes. So this means that a point A from a point B that distance is almost indistinguishable from a point A and point C. And this can lead to some problems specifically in machine learning where slightly different sets of data can lead to different model behavior. And hence the curse of dimensionality can also lead to overfitting in machine learning models such as k-means, k-nearest neighbors, and decision trees among others.