Backward pass (loss.backward) in PyTorch - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the purpose of the backward pass in PyTorch?

The backward pass computes gradients of the loss with respect to model parameters. It helps the model learn by showing how to adjust weights to reduce errors.

beginner

What does the method loss.backward() do in PyTorch?

It calculates the gradients of the loss tensor with respect to all tensors that have requires_grad=True. These gradients are stored in the .grad attribute of each tensor.

intermediate

Why do we need to call optimizer.zero_grad() before loss.backward()?

Because PyTorch accumulates gradients by default, calling optimizer.zero_grad() clears old gradients. This prevents mixing gradients from multiple backward passes.

beginner

What happens if you forget to call loss.backward() during training?

No gradients will be computed, so the optimizer cannot update the model weights. The model will not learn or improve.

intermediate

How does PyTorch know which operations to track for gradient computation?

PyTorch builds a computation graph dynamically during the forward pass. It tracks operations on tensors with requires_grad=True to compute gradients during the backward pass.

What does loss.backward() compute?

A The optimizer step

B Gradients of loss with respect to model parameters

C The updated model weights

D The loss value itself

loss.backward() calculates gradients needed for updating model weights.

Before calling loss.backward(), why do we call optimizer.zero_grad()?

A To clear old gradients so they don't accumulate

B To compute the loss value

C To update the model weights

D To save the model

Gradients accumulate by default, so clearing them prevents mixing updates.

If a tensor has requires_grad=False, what happens during loss.backward()?

A Gradient is computed normally

B The tensor is updated automatically

C An error occurs

D No gradient is computed for that tensor

Only tensors with requires_grad=True get gradients computed.

What is stored in the .grad attribute after loss.backward()?

A The updated tensor value

B The loss value

C The gradient of the loss with respect to the tensor

D The optimizer state

.grad holds the computed gradients for each parameter.

What does PyTorch use to track operations for gradient calculation?

A A dynamic computation graph

B A static computation graph

C A database

D A configuration file

PyTorch builds a dynamic graph during the forward pass to track operations.

Explain in your own words what happens during the backward pass when you call loss.backward() in PyTorch.

Think about how PyTorch figures out how to change weights to reduce loss.

Why is it important to call optimizer.zero_grad() before loss.backward() in a training loop?

Consider what happens if gradients from previous steps mix with current ones.

Practice

1. What does the loss.backward() function do in PyTorch?

easy

A. It computes gradients of the loss with respect to model parameters.

B. It updates the model weights directly.

C. It calculates the loss value from predictions and targets.

D. It initializes the model parameters.

Solution

1. Step 1: Understand the role of loss.backward()

   This function computes gradients needed for updating model weights by backpropagation.

2. Step 2: Differentiate from other functions

   Updating weights is done by the optimizer, not loss.backward(). Calculating loss or initializing parameters are separate steps.

3. Final Answer:

   It computes gradients of the loss with respect to model parameters. -> Option A

4. Quick Check:

   Backward pass = compute gradients [OK]

Hint: Remember: backward() computes gradients, optimizer updates weights [OK]

Common Mistakes:

• Confusing backward() with optimizer step
• Thinking backward() calculates loss value
• Assuming backward() initializes parameters

2. Which of the following is the correct way to call the backward pass on a loss tensor named loss in PyTorch?

easy

A. loss.backwards()

B. loss.backward()

C. loss.backwarded()

D. loss.backward_pass()

Solution

1. Step 1: Recall PyTorch syntax for backward pass

   The correct method to compute gradients is loss.backward().

2. Step 2: Identify incorrect method names

   Other options are invalid method names and will cause errors.

3. Final Answer:

   loss.backward() -> Option B

4. Quick Check:

   Correct method name = backward() [OK]

Hint: Use exact method name: backward() with no extra letters [OK]

Common Mistakes:

• Adding extra letters or suffixes to backward
• Using plural form backwards
• Confusing method name with optimizer step

3. Consider the following PyTorch code snippet:

import torch
x = torch.tensor(2.0, requires_grad=True)
y = x ** 3
loss = y
loss.backward()
print(x.grad)

What will be the output of print(x.grad)?

medium

A. tensor(0.)

B. tensor(8.)

C. tensor(6.)

D. tensor(12.)

Solution

1. Step 1: Compute the derivative of y = x^3 at x=2

   The derivative dy/dx = 3 * x^2. At x=2, this is 3 * 4 = 12.

2. Step 2: Understand backward() computes this gradient

   Calling loss.backward() computes and stores this gradient in x.grad.

3. Final Answer:

   tensor(12.) -> Option D

4. Quick Check:

   Derivative of x^3 at 2 = 12 [OK]

Hint: Derivative of x^3 is 3x^2; plug in x=2 [OK]

Common Mistakes:

• Calculating derivative incorrectly as 8 or 6
• Forgetting to set requires_grad=True
• Not calling loss.backward() before accessing grad

4. What is the error in the following PyTorch code snippet?

import torch
x = torch.tensor(3.0, requires_grad=True)
y = x * 2
loss = y
# Missing backward call
print(x.grad)

medium

A. x.grad will be zero tensor automatically.

B. SyntaxError because backward() is missing.

C. x.grad will be None because backward() was not called.

D. RuntimeError because loss is not scalar.

Solution

1. Step 1: Identify missing backward() call

   The code does not call loss.backward(), so gradients are not computed.

2. Step 2: Understand gradient attribute behavior

   Without backward(), x.grad remains None, not zero or error.

3. Final Answer:

   x.grad will be None because backward() was not called. -> Option C

4. Quick Check:

   No backward() means no gradients [OK]

Hint: Always call loss.backward() before checking .grad [OK]

Common Mistakes:

• Expecting zero tensor instead of None
• Thinking missing backward() causes syntax error
• Assuming RuntimeError for non-scalar loss here

5. You have a model with parameters and a loss computed. You call loss.backward() twice in a row without zeroing gradients in between. What will happen to the gradients stored in the parameters?

hard

A. Gradients will accumulate, doubling the values from one backward call.

B. Gradients will reset to zero before the second backward call.

C. An error will be raised because backward() cannot be called twice.

D. Gradients will be overwritten by the second backward call.

Solution

1. Step 1: Understand gradient accumulation in PyTorch

   By default, gradients accumulate in the .grad attribute when backward() is called multiple times.

2. Step 2: Effect of calling backward() twice without zeroing

   Calling backward() twice adds gradients twice, effectively doubling them.

3. Final Answer:

   Gradients will accumulate, doubling the values from one backward call. -> Option A

4. Quick Check:

   Multiple backward calls add gradients [OK]

Hint: Call optimizer.zero_grad() to reset gradients before backward() [OK]

Common Mistakes:

• Assuming gradients reset automatically
• Expecting an error on second backward call
• Thinking gradients overwrite instead of accumulate