Manhattan distance in Artificial Intelligence is the sum of absolute distance between two points across all the axes. Suppose there are two points (p1, q1), (p2, q2) in two dimensions, then Manhattan distance between two points is
Manhattan distance = | p1-q1 | + | p2-q2 |
Manhattan distance is to calculate the similarity between data points or observations for classification problems and clustering.
If you want to get started in Artificial Intelligence with TensorFlow, I would recommend enrolling in this Artificial Intelligence training by Intellipaat.