Editing Statistical Learning Theory (section)

== <span style="color: #FFFFFF;">Applying</span> ==
'''Modeling 'Overfitting' (Polynomial Regression Logic):'''
<syntaxhighlight lang="python">
import numpy as np

def calculate_error(y_true, y_pred):
    return np.mean((y_true - y_pred)**2)

# True Pattern: y = x + noise
x = np.array([1, 2, 3, 4, 5])
y = x + np.random.normal(0, 0.5, 5)

# Simple Model (Linear): y_pred = x
y_simple = x
# Complex Model (Overfit): hits every point exactly
y_complex = y

print(f"Training Error (Simple):  {calculate_error(y, y_simple):.3f}")
print(f"Training Error (Complex): {calculate_error(y, y_complex):.3f}")
# The complex model looks better on paper (0 error!), but it 
# will fail miserably on the NEXT data point.
</syntaxhighlight>

; Common Learning Algorithms
: '''Linear Regression''' → Predicting a continuous number (e.g., house prices).
: '''Logistic Regression''' → Predicting a category (e.g., 'Will buy' or 'Won't buy').
: '''K-Means Clustering''' → Grouping data points into 'K' similar clusters.
: '''Random Forests''' → Combining the predictions of hundreds of 'Decision Trees' to get a more accurate result.
: '''Support Vector Machines (SVM)''' → Finding the "widest gap" between categories.
</div>

<div style="background-color: #8B4500; color: #FFFFFF; padding: 20px; border-radius: 8px; margin-bottom: 15px;">