MixUp strategy in Computer Vision - Model Metrics & Evaluation

Metrics & Evaluation - MixUp strategy

Which metric matters for MixUp strategy and WHY

MixUp is a data augmentation method that blends images and their labels. It helps models generalize better. Because it changes training data, the key metrics to watch are validation accuracy and validation loss. These show if the model learns well on new, unseen data. Also, robustness metrics like accuracy on noisy or adversarial examples matter, since MixUp aims to improve model stability.

Confusion matrix example with MixUp

    Actual \ Predicted | Cat | Dog | Bird
    -------------------|-----|-----|-----
    Cat                | 45  | 3   | 2
    Dog                | 4   | 43  | 3
    Bird               | 1   | 2   | 47

    Total samples = 150
    TP (example for Cat) = 45
    FP (Cat predicted but not Cat) = 4 + 1 = 5
    FN (Cat actual but not predicted) = 3 + 2 = 5
    Precision (Cat) = 45 / (45 + 5) = 0.9
    Recall (Cat) = 45 / (45 + 5) = 0.9
    

This matrix shows balanced precision and recall, indicating MixUp helped the model generalize well across classes.

Precision vs Recall tradeoff with MixUp

MixUp encourages smoother decision boundaries, which can improve both precision and recall. But sometimes, increasing recall (catching more true positives) may lower precision (more false positives). For example:

• High precision, low recall: Model is very sure but misses some true cases.
• High recall, low precision: Model catches most true cases but also many wrong ones.

MixUp helps balance this tradeoff by making the model less confident on ambiguous samples, improving overall robustness.

What good vs bad metric values look like for MixUp

Good:

• Validation accuracy steadily improves or stays stable compared to baseline.
• Validation loss decreases smoothly without sudden jumps.
• Precision and recall are balanced and high (e.g., above 85%).
• Model performs well on noisy or mixed inputs.

Bad:

• Validation accuracy drops or fluctuates wildly.
• Validation loss is high or unstable.
• Precision or recall is very low, showing poor generalization.
• Model fails on augmented or mixed samples, indicating overfitting.

Common pitfalls in metrics when using MixUp

• Accuracy paradox: Accuracy may look good but model might not handle real mixed data well.
• Data leakage: Mixing training and validation data accidentally can inflate metrics.
• Overfitting signs: Training loss very low but validation loss high means model memorizes instead of generalizing.
• Ignoring robustness: Only checking accuracy on clean data misses MixUp benefits on noisy inputs.

Self-check question

Your model trained with MixUp has 98% accuracy but only 12% recall on a rare class. Is it good for production? Why or why not?

Answer: No, it is not good. High accuracy can be misleading if the rare class is small. Low recall means the model misses most true cases of that class, which can be critical depending on the task. You should improve recall to catch more true positives, even if accuracy drops slightly.

Key Result

MixUp improves model robustness by balancing precision and recall, shown by stable validation accuracy and loss.

Practice

1. What is the main purpose of the MixUp strategy in computer vision?

easy

A. To blend images and labels to create smoother decision boundaries

B. To increase the size of the neural network

C. To reduce the image resolution for faster training

D. To remove noisy images from the dataset

Solution

1. Step 1: Understand MixUp's function

   MixUp combines two images and their labels to create new training samples.

2. Step 2: Identify the goal of blending

   This blending helps the model learn smoother decision boundaries and generalize better.

3. Final Answer:

   To blend images and labels to create smoother decision boundaries -> Option A

4. Quick Check:

   MixUp = blending images and labels [OK]

Hint: MixUp mixes images and labels to smooth learning [OK]

Common Mistakes:

• Thinking MixUp increases model size
• Confusing MixUp with image resizing
• Believing MixUp removes data

2. Which of the following is the correct way to generate a mixed image x_mix using MixUp with images x1, x2 and mixing coefficient lambda?

easy

A. x_mix = (1 - lambda) * x1 + lambda * x2

B. x_mix = x1 + x2

C. x_mix = lambda * (x1 + x2)

D. x_mix = lambda * x1 + (1 - lambda) * x2

Solution

1. Step 1: Recall MixUp formula for images

   MixUp creates a weighted sum: mixed image = (1 - lambda) * first image + lambda * second image.

2. Step 2: Match formula to options

   x_mix = (1 - lambda) * x1 + lambda * x2 matches this formula exactly.

3. Final Answer:

   x_mix = (1 - lambda) * x1 + lambda * x2 -> Option A

4. Quick Check:

   MixUp image = weighted sum with lambda [OK]

Hint: MixUp image = (1-lambda) * first + lambda * second [OK]

Common Mistakes:

• Swapping lambda and (1 - lambda) incorrectly
• Adding images without weights
• Multiplying sum by lambda instead of weighting separately

3. Given the following code snippet using MixUp, what will be the shape of mixed_labels?

import torch
labels1 = torch.tensor([[1, 0, 0], [0, 1, 0]])
labels2 = torch.tensor([[0, 0, 1], [1, 0, 0]])
lambda_ = 0.7
mixed_labels = lambda_ * labels1 + (1 - lambda_) * labels2
print(mixed_labels.shape)

medium

A. torch.Size([2])

B. torch.Size([3, 2])

C. torch.Size([2, 3])

D. torch.Size([3])

Solution

1. Step 1: Check input label shapes

   labels1 and labels2 are tensors with shape [2, 3] (2 samples, 3 classes).

2. Step 2: Mixing labels keeps shape

   MixUp combines labels element-wise, so mixed_labels shape remains [2, 3].

3. Final Answer:

   torch.Size([2, 3]) -> Option C

4. Quick Check:

   Mixed labels shape = input labels shape [OK]

Hint: MixUp label shape matches original labels [OK]

Common Mistakes:

• Confusing batch and class dimensions
• Assuming shape changes after mixing
• Misreading tensor dimensions

4. Identify the error in this MixUp implementation snippet:

def mixup_data(x1, x2, y1, y2, alpha=1.0):
    lam = np.random.beta(alpha, alpha)
    x_mix = lam * x1 + (1 - lam) * x2
    y_mix = lam * y1 + lam * y2
    return x_mix, y_mix

medium

A. The beta distribution parameters are incorrect

B. The function should return only mixed images

C. The images should not be mixed with weights

D. The label mixing formula should use (1 - lam) for y2, not lam

Solution

1. Step 1: Review label mixing formula

   Labels must be mixed as y_mix = lam * y1 + (1 - lam) * y2 to match image mixing.

2. Step 2: Identify mismatch in code

   The code incorrectly uses lam * y2 instead of (1 - lam) * y2.

3. Final Answer:

   The label mixing formula should use (1 - lam) for y2, not lam -> Option D

4. Quick Check:

   Label weights must sum to 1 [OK]

Hint: Label weights must add to 1 in MixUp [OK]

Common Mistakes:

• Using same weight for both labels
• Ignoring beta distribution for lambda
• Returning incomplete outputs

5. You want to apply MixUp to a dataset with 10,000 images and 5 classes. If you set the MixUp parameter alpha very high (e.g., 100), what is the likely effect on training?

hard

A. The model will train faster due to clearer labels

B. The mixed images will be almost equal blends, possibly making labels ambiguous

C. The mixed images will be mostly original images with little blending

D. The model will ignore MixUp and train normally

Solution

1. Step 1: Understand beta distribution with high alpha

   High alpha in beta distribution centers lambda near 0.5, causing equal blending.

2. Step 2: Consider impact on labels and images

   Equal blending creates ambiguous images and soft labels, which may confuse the model.

3. Final Answer:

   The mixed images will be almost equal blends, possibly making labels ambiguous -> Option B

4. Quick Check:

   High alpha = balanced blends, ambiguous labels [OK]

Hint: High alpha means balanced mixes, can confuse model [OK]

Common Mistakes:

• Thinking high alpha means less mixing
• Assuming faster training with high alpha
• Believing MixUp is ignored with extreme alpha