Machine Learning Fundamentals Cheat Sheet

Machine learning (ML) enables systems to learn from data without explicit programming. Key types include supervised, unsupervised, and reinforcement learning, each suited for different problem types like classification, regression, and decision-making.

Core Principles

Supervised Learning: Learns from labeled data (input-output pairs) to predict outcomes.
Unsupervised Learning: Discovers patterns in unlabeled data, such as clustering or dimensionality reduction.
Reinforcement Learning: Learns through trial and error, receiving rewards or penalties for actions.
Classification: Predicts a discrete category or class label (e.g., spam or not spam).
Regression: Predicts a continuous numerical value (e.g., house price or temperature).
Real-world applications span from image recognition to personalized recommendations.

Key Terms

Supervised Learning: Learning from data with known outcomes or labels.
Unsupervised Learning: Learning from data without predefined labels or outcomes.
Reinforcement Learning: Learning through interaction with an environment, based on rewards and punishments.
Classification: A supervised learning task to assign data points to predefined categories.
Regression: A supervised learning task to predict a continuous numerical output.

Real World Examples

Email Spam Detection: Classification (Supervised Learning)
House Price Prediction: Regression (Supervised Learning)
Customer Segmentation: Clustering (Unsupervised Learning)
Game Playing AI (e.g., Chess): Reinforcement Learning
Image Recognition: Classification (Supervised Learning)