### Description  highlights the three major categories of Machine Learning, along with a brief description, common algorithms, and typical use cases. | **Category** | **Key Idea** | **Common Algorithms** | **Typical Use Cases** | | | -------------------------- | ----------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------- | --- | | **Supervised Learning** | Training a model on labeled data (features + known target outputs) | - Linear Regression<br>- Logistic Regression<br>- Decision Trees<br>- Random Forests<br>- Support Vector Machines (SVM)<br>- Neural Networks (e.g., MLP) | - Spam filtering<br>- Image classification<br>- Stock price prediction<br>- Speech recognition | | | **Unsupervised Learning** | Finding patterns or structures in unlabeled data (only features, no labels) | - K-means Clustering<br>- Hierarchical Clustering<br>- DBSCAN<br>- Principal Component Analysis (PCA)<br>- Autoencoders | - Customer segmentation<br>- Anomaly detection<br>- [[Dimensionality reduction]]<br>- Recommender systems (item similarity) | | | **Reinforcement Learning** | An agent learns optimal actions through rewards or penalties by interacting with an environment | - Q-Learning<br>- Deep Q-Network (DQN)<br>- Policy Gradients<br>- Proximal Policy Optimization (PPO) | - Robotics<br>- Game playing (e.g., AlphaGo)<br>- Resource allocation<br>- Autonomous driving | |