Gradient and Backpropagation

2 min

Objective: understand how a final error becomes a local correction.

The gradient gathers the derivatives of the loss with respect to the parameters. Its component ∂L/∂w_i indicates how a small increase in weight w_i would affect the loss.

The chain rule

If L depends on a, which depends on z, which depends on w:

text
∂L/∂w = ∂L/∂a × ∂a/∂z × ∂z/∂w

Backpropagation applies this rule efficiently from the loss down to every parameter of the graph. It is not the optimizer: it computes the gradients; the optimizer then decides the update.

python
optimizer.zero_grad()  # clear previous gradients
logits = model(x)
loss = criterion(logits, y)
loss.backward()        # compute gradients
optimizer.step()       # update parameters

Preview — the rest of the lesson is for enrolled readers.

Already enrolled with a code?

Your access is tied to your account, not to this link. Sign in with the same email you used in class: your course is waiting, no need to enter the code again.

Sign inNo account yet? Create one
This lesson is part of the “Training the Network” module

The first modules of the course are open to everyone. For the rest you have three options: buy this course once and for all, subscribe, or enter the code handed out in class.

Are you a student on this course?

The code is tied to your account: sign in or create an account and it will be applied automatically when you come back.

No account yet? Create one