Learning rate differential in PyTorch - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is a learning rate in machine learning?

The learning rate is a small number that controls how much the model changes its knowledge each time it learns from data. Think of it like the speed of learning.

beginner

What does 'learning rate differential' mean?

Learning rate differential means using different learning rates for different parts of a model. Some parts learn faster, others slower, like giving more attention to some tasks.

intermediate

Why use different learning rates for different layers in a neural network?

Because some layers may need bigger changes to learn new things, while others need smaller changes to keep what they already know. This helps the model learn better and faster.

intermediate

How do you set different learning rates in PyTorch?

You can pass a list of dictionaries to the optimizer, each with a 'params' key for the model part and a 'lr' key for its learning rate. This tells PyTorch to update each part with its own speed.

intermediate

What is a practical example of learning rate differential?

When fine-tuning a pre-trained model, you might use a small learning rate for the old layers to keep their knowledge, and a bigger learning rate for new layers to learn fast.

What does a higher learning rate do?

A Makes the model learn faster but risks missing details

B Makes the model learn slower and more carefully

C Stops the model from learning

D Has no effect on learning

A higher learning rate means bigger steps in learning, which can speed up training but might skip over important details.

In PyTorch, how do you apply different learning rates to different layers?

A By setting a global learning rate only

B By passing a list of parameter groups with different 'lr' values to the optimizer

C By using different optimizers for each layer

D By changing the learning rate after each epoch manually

PyTorch allows setting different learning rates by passing parameter groups with their own 'lr' values to the optimizer.

Why might you want a smaller learning rate for pre-trained layers?

A To stop them from updating completely

B To make them learn faster

C To keep their learned knowledge stable

D To reset their weights

A smaller learning rate helps preserve the useful knowledge already learned in pre-trained layers.

What is a risk of using too large a learning rate?

A Model might not learn well and jump around

B Model will learn perfectly

C Training will be very slow

D Model will ignore the data

Too large a learning rate can cause the model to miss the best solution by jumping too much.

Learning rate differential is especially useful in which scenario?

A When not using an optimizer

B Training a model from scratch with one layer

C Using a fixed learning rate for all layers

D Fine-tuning a pre-trained model

Fine-tuning benefits from different learning rates to adjust new and old layers properly.

Explain what learning rate differential is and why it helps in training neural networks.

Think about how some parts of the model might need to learn slower or faster.

Describe how to implement learning rate differential in PyTorch with code.

Focus on how the optimizer receives different learning rates.

Practice

1. What does learning rate differential mean in PyTorch training?

easy

A. Changing the learning rate randomly during training

B. Setting different learning rates for different parts of a model

C. Using the same learning rate for the entire model

D. Freezing all model layers during training

Solution

1. Step 1: Understand learning rate concept

   The learning rate controls how fast a model updates its knowledge during training.

2. Step 2: Define learning rate differential

   Learning rate differential means assigning different learning rates to different parts of the model to control their update speed.

3. Final Answer:

   Setting different learning rates for different parts of a model -> Option B

4. Quick Check:

   Learning rate differential = Different rates per model part [OK]

Hint: Different parts can learn at different speeds [OK]

Common Mistakes:

• Thinking learning rate is always the same for all layers
• Confusing learning rate differential with random rate changes
• Believing freezing layers means changing learning rate

2. Which PyTorch code snippet correctly sets different learning rates for two parameter groups?

easy

A. optimizer = torch.optim.SGD(model.parameters(), lr=0.01, lr2=0.001)

B. optimizer = torch.optim.SGD(model.parameters(), lr=[0.01, 0.001])

C. optimizer = torch.optim.SGD([{'params': model.layer1.parameters(), 'lr': 0.01}, {'params': model.layer2.parameters(), 'lr': 0.001}], momentum=0.9)

D. optimizer = torch.optim.SGD([model.layer1, model.layer2], lr=0.01)

Solution

1. Step 1: Check PyTorch optimizer syntax for param groups

   PyTorch allows passing a list of dicts with 'params' and 'lr' keys to set different learning rates.

2. Step 2: Identify correct syntax

   optimizer = torch.optim.SGD([{'params': model.layer1.parameters(), 'lr': 0.01}, {'params': model.layer2.parameters(), 'lr': 0.001}], momentum=0.9) correctly uses a list of dicts with separate learning rates for layer1 and layer2 parameters.

3. Final Answer:

   optimizer = torch.optim.SGD([{'params': model.layer1.parameters(), 'lr': 0.01}, {'params': model.layer2.parameters(), 'lr': 0.001}], momentum=0.9) -> Option C

4. Quick Check:

   Param groups with separate 'lr' keys = Correct syntax [OK]

Hint: Use list of dicts with 'params' and 'lr' keys [OK]

Common Mistakes:

• Passing lr as a list directly to optimizer
• Using unknown keyword like lr2
• Passing layers instead of parameters

3. Given this code, what is the learning rate for model.layer2 during training?

optimizer = torch.optim.Adam([
  {'params': model.layer1.parameters(), 'lr': 0.005},
  {'params': model.layer2.parameters(), 'lr': 0.0005}
])

medium

A. 0.0005

B. 0.05

C. 0.0055

D. 0.005

Solution

1. Step 1: Identify learning rates assigned to each layer

   Layer1 has lr=0.005, Layer2 has lr=0.0005 as per the optimizer param groups.

2. Step 2: Find learning rate for model.layer2

   From the second dict, model.layer2.parameters() uses lr=0.0005.

3. Final Answer:

   0.0005 -> Option A

4. Quick Check:

   Layer2 lr = 0.0005 from param groups [OK]

Hint: Check param group with layer2 parameters [OK]

Common Mistakes:

• Adding learning rates instead of selecting correct one
• Confusing layer1 lr with layer2 lr
• Assuming default lr overrides param groups

4. Identify the error in this PyTorch optimizer setup for learning rate differential:

optimizer = torch.optim.SGD([
  {'params': model.layer1.parameters(), 'lr': 0.01},
  {'params': model.layer2.parameters()}
], lr=0.001)

medium

A. Missing learning rate for second param group causes error

B. Using lr=0.001 outside param groups is invalid

C. Parameters should be passed as model.layer1, not model.layer1.parameters()

D. SGD optimizer does not support param groups

Solution

1. Step 1: Review param groups and learning rates

   First param group has lr=0.01, second param group has no lr specified.

2. Step 2: Understand default lr behavior

   When param groups are used, each group should have lr or optimizer's lr applies. Here, lr=0.001 is passed but second group lacks explicit lr, causing confusion.

3. Final Answer:

   Missing learning rate for second param group causes error -> Option A

4. Quick Check:

   All param groups need lr or default applies [OK]

Hint: Each param group must have lr or rely on optimizer lr [OK]

Common Mistakes:

• Assuming optimizer lr applies to all param groups automatically
• Passing parameters instead of parameter iterators
• Believing SGD can't use param groups

5. You want to fine-tune a pretrained model by training only the last layer fast and freezing the rest. Which learning rate setup is best?

hard

A. Set same lr=0.01 for all layers

B. Freeze last layer and train others with lr=0.01

C. Set lr=0.01 for all layers except last layer with lr=0

D. Set lr=0 for all layers except last layer with lr=0.01

Solution

1. Step 1: Understand freezing and learning rate

   Freezing means no updates, which can be done by setting lr=0 or disabling gradients.

2. Step 2: Apply learning rate differential for fine-tuning

   Set lr=0 for frozen layers to prevent updates, and higher lr for last layer to train it fast.

3. Final Answer:

   Set lr=0 for all layers except last layer with lr=0.01 -> Option D

4. Quick Check:

   Freeze layers = lr 0, train last layer fast [OK]

Hint: Freeze layers by lr=0, train last layer with higher lr [OK]

Common Mistakes:

• Using same learning rate for all layers when freezing
• Freezing last layer instead of others
• Not setting lr=0 for frozen layers