Explore Library
Code QuizIntermediate

Joint and Marginal Probability

Identify the axis error when marginalizing a joint probability table.

Codepython
import numpy as np

# Joint distribution P(X, Y); rows = X, cols = Y
joint = np.array([[0.1, 0.2],
                  [0.3, 0.4]])

# Marginal P(X): sum over Y for each value of X
p_x = joint.sum(axis=0)
print(p_x)  # expected one probability per X value

What is the bug in computing the marginal P(X)?