Machine Learning
Data Pre-processing & Feature Engineering
Cleaning, transforming, and selecting useful features.
Ensures data quality and improves model performance.
Critical first step before applying any ML algorithm.
Supervised Learning
Learning from labelled data to make predictions.
Includes regression and classification problems.
Forms the foundation of most practical ML systems.
Unsupervised Learning
Discovering hidden patterns in unlabelled data.
Includes clustering and dimensionality reduction.
Useful for exploration and representation learning.
Model Evaluation & Validation
Measuring model performance and generalization.
Uses metrics, cross-validation, and test sets.
Prevents overfitting and underfitting.
Optimization Techniques
Methods to minimize loss functions efficiently.
Includes gradient descent and its variants.
Essential for training ML and neural models.
Regularization & Bias-Variance Trade-off
Techniques to control model complexity.
Improves generalization on unseen data.
Key to building robust ML systems.
