Mutual Information I(X;Y)

Drag the sliders to shape a 2D joint distribution. Watch the joint heatmap, marginals, and I(X;Y) = H(X) + H(Y) − H(X,Y) update live.

—

H(X) bits

—

H(Y) bits

—

H(X,Y) bits

—

I(X;Y) bits

Correlation ρ 0.00

Distribution bins 6

Spread σ 0.50

I(X;Y) = 0 means X and Y are independent. Maximum I(X;Y) = min(H(X), H(Y)) when X and Y are perfectly correlated.