Bias detection and mitigation in ML Python - Model Pipeline Trace

Model Pipeline - Bias detection and mitigation

This pipeline shows how we detect bias in data and reduce it to make fairer predictions. We start with raw data, check for bias, adjust the data or model, train the model, and then check if bias is reduced.

Data Flow - 6 Stages

1Raw Data Collection

1000 rows x 6 columns→Collect data including features and sensitive attribute (e.g., gender)→1000 rows x 6 columns

Rows with features like age, income, education, gender, and target label

↓

2Bias Detection

1000 rows x 6 columns→Calculate bias metrics like demographic parity difference→Bias score: 0.25 (indicating bias)

Demographic parity difference = 0.25 means groups show disparity

↓

3Bias Mitigation - Reweighing

1000 rows x 6 columns→Assign weights to samples to balance groups→1000 rows x 6 columns with weights

Samples from underrepresented group get higher weights

↓

4Train/Test Split

1000 rows x 6 columns with weights→Split data into training (80%) and testing (20%) sets→Training: 800 rows x 6 columns with weights, Testing: 200 rows x 6 columns without weights

Training set has 800 samples with weights

↓

5Model Training

800 rows x 6 columns with weights→Train weighted logistic regression model→Trained model

Model learns to predict target with bias mitigation

↓

6Evaluation

200 rows x 6 columns→Calculate accuracy and bias metrics on test set→Accuracy: 0.82, Bias score: 0.05

Bias score reduced from 0.25 to 0.05 after mitigation

Training Trace - Epoch by Epoch

Loss
0.7 |****
0.6 |****
0.5 |****
0.4 |****
0.3 |****
    +------------
     1 2 3 4 5 Epochs

Epoch	Loss ↓	Accuracy ↑	Observation
1	0.65	0.60	Initial training with high loss and low accuracy
2	0.50	0.70	Loss decreased, accuracy improved
3	0.40	0.75	Model learning patterns, bias mitigation helping
4	0.35	0.80	Loss continues to decrease, accuracy rising
5	0.30	0.82	Training converging with good accuracy and lower bias

Prediction Trace - 6 Layers

Layer 1: Input Sample

Layer 2: Feature Encoding

Layer 3: Model Input

Layer 4: Logistic Regression Model

Layer 5: Sigmoid Activation

Layer 6: Final Prediction

Model Quiz - 3 Questions

What does a bias score of 0.25 before mitigation indicate?

A The model accuracy is 25%

B The model has perfect fairness

C The model favors one group over another

D The data has no sensitive attributes

Key Insight

Detecting bias early and applying mitigation like reweighing helps train fairer models without sacrificing accuracy. Monitoring bias metrics alongside accuracy ensures balanced predictions.

Practice

1. What is the main goal of bias detection in machine learning models?

easy

A. To find unfair treatment of certain groups by the model

B. To increase the speed of model training

C. To reduce the size of the dataset

D. To improve the model's accuracy on all data

Solution

1. Step 1: Understand bias detection purpose

   Bias detection aims to identify if a model treats some groups unfairly compared to others.

2. Step 2: Compare options to goal

   Only To find unfair treatment of certain groups by the model matches this goal; others focus on speed, size, or accuracy, which are unrelated.

3. Final Answer:

   To find unfair treatment of certain groups by the model -> Option A

4. Quick Check:

   Bias detection = find unfair treatment [OK]

Hint: Bias detection finds unfairness in model decisions [OK]

Common Mistakes:

• Confusing bias detection with improving accuracy
• Thinking bias detection speeds up training
• Mixing bias detection with data size reduction

2. Which of the following is a common technique to mitigate bias in training data?

easy

A. Oversampling underrepresented groups

B. Reducing the number of features

C. Increasing the learning rate

D. Using dropout layers in neural networks

Solution

1. Step 1: Identify bias mitigation methods

   Oversampling underrepresented groups balances data, reducing bias.

2. Step 2: Evaluate other options

   Reducing features, increasing learning rate, and dropout affect model complexity or training, not bias directly.

3. Final Answer:

   Oversampling underrepresented groups -> Option A

4. Quick Check:

   Bias mitigation = oversampling minority data [OK]

Hint: Balance data by oversampling small groups [OK]

Common Mistakes:

• Confusing feature reduction with bias mitigation
• Thinking learning rate affects bias
• Mixing dropout with bias correction

3. Consider this Python code snippet for bias mitigation:

from sklearn.utils import resample

# Original data
X = [[1], [2], [3], [4]]
y = [0, 0, 1, 1]

# Minority class samples
X_minority = [x for x, label in zip(X, y) if label == 1]
y_minority = [1, 1]

# Oversample minority class
X_upsampled, y_upsampled = resample(X_minority, y_minority, replace=True, n_samples=4, random_state=42)

print(len(X_upsampled), len(y_upsampled))

What will be the output of the print statement?

medium

A. 2 2

B. 4 4

C. 8 8

D. Error due to wrong resample usage

Solution

1. Step 1: Understand resample parameters

   resample duplicates minority samples to reach n_samples=4, so output lists have length 4.

2. Step 2: Check print output

   Both X_upsampled and y_upsampled have length 4, so print outputs '4 4'.

3. Final Answer:

   4 4 -> Option B

4. Quick Check:

   Oversampling size = 4 [OK]

Hint: resample with n_samples sets output length [OK]

Common Mistakes:

• Assuming output length stays original size
• Thinking resample doubles data always
• Expecting error from resample usage

4. You have a model that shows bias against a group. The code below tries to fix it but causes an error:

def mitigate_bias(data, labels):
    # Remove samples from majority group
    filtered_data = [d for d, l in zip(data, labels) if l != 'majority']
    filtered_labels = [l for l in labels if l != 'majority']
    return filtered_data, filtered_labels

X = [1, 2, 3, 4]
y = ['majority', 'minority', 'majority', 'minority']

X_new, y_new = mitigate_bias(X, y)
print(len(X_new), len(y_new))

What is the error in this code?

medium

A. Function returns wrong variable names

B. Syntax error in list comprehension

C. No error, code runs correctly

D. filtered_labels length does not match filtered_data length

Solution

1. Step 1: Analyze the filtering logic

   filtered_data is filtered using zip(data, labels) with condition l != 'majority', selecting elements at indices 1 and 3. filtered_labels is filtered directly from labels with the same condition, resulting in elements at indices 1 and 3. However, filtered_data and filtered_labels are filtered differently: filtered_data uses zip to filter data and labels together, but filtered_labels filters labels alone, which can cause length mismatch if data and labels are not aligned.

2. Step 2: Check lengths

   In this specific example, both filtered_data and filtered_labels have length 2, but this is coincidental. The approach is error-prone and can cause mismatches in other cases.

3. Step 3: Correct approach

   Both filtered_data and filtered_labels should be filtered using zip together to ensure matching lengths.

4. Final Answer:

   filtered_labels length does not match filtered_data length -> Option D

5. Quick Check:

   Separate filtering risks length mismatch [OK]

Hint: Filter data and labels together using zip to keep lengths equal [OK]

Common Mistakes:

• Mistakenly thinking separate comprehensions cause length mismatch
• Believing zip and direct label filter select differently
• Assuming the code has a bug without tracing execution

5. You want to reduce bias in a model predicting loan approvals. You have data with a sensitive attribute 'gender'. Which combined approach best mitigates bias while keeping model performance?

hard

A. Increase model complexity to capture more patterns

B. Remove the 'gender' attribute and train the model normally

C. Use reweighting to balance training samples and add fairness constraints during model training

D. Only oversample the majority group to increase data size

Solution

1. Step 1: Understand bias mitigation strategies

   Reweighting balances data impact; fairness constraints guide model to treat groups fairly.

2. Step 2: Evaluate other options

   Removing 'gender' may hide bias but not fix it; increasing complexity can worsen bias; oversampling majority increases imbalance.

3. Final Answer:

   Use reweighting to balance training samples and add fairness constraints during model training -> Option C

4. Quick Check:

   Combine reweighting + fairness constraints = best bias mitigation [OK]

Hint: Combine data balancing and fairness rules for best results [OK]

Common Mistakes:

• Removing sensitive attribute hides but does not fix bias
• Increasing complexity ignores bias issues
• Oversampling majority worsens bias