Computing Directional Derivatives Using the Gradient
You'll learn: How to quickly compute the rate of change in any direction using the gradient vector and a dot product.
The Key Formula
In the previous lesson, you learned what a directional derivative is—the rate of change of a function in a specific direction. Now you'll learn the powerful shortcut for computing it.
Instead of using the limit definition every time, we have a remarkably efficient formula:
D_u f = ∇f · u
where:
- D_u f is the directional derivative in direction u
- ∇f is the gradient vector (which you already know how to compute)
- u is a unit direction vector (magnitude = 1)
- The · represents the dot product
Why This Works
Think of the gradient ∇f as containing all the "raw" rate-of-change information at a point. The dot product projects this onto your chosen direction u, extracting exactly the component of change along that direction. It's like asking: "How much of the gradient's push aligns with where I want to go?"
Example
Suppose f(x, y) = x² + 3xy and you want the directional derivative at (1, 2) in the direction of v = ⟨3, 4⟩.
Step 1: Compute ∇f = ⟨2x + 3y, 3x⟩
At (1, 2): ∇f = ⟨2(1) + 3(2), 3(1)⟩ = ⟨8, 3⟩
Step 2: Normalize v to a unit vector:
|v| = √(3² + 4²) = 5
u = ⟨3/5, 4/5⟩
Step 3: Take the dot product:
D_u f = ⟨8, 3⟩ · ⟨3/5, 4/5⟩ = (8)(3/5) + (3)(4/5) = 24/5 + 12/5 = 36/5
Key Takeaway: The directional derivative in direction u equals the dot product ∇f · u, where u must be a unit vector—this replaces tedious limit calculations with simple vector arithmetic.