Machine learning (ML) is a branch of AI that lets computers learn from data instead of being explicitly programmed with rules.
ML algoritms: 1) Supervised learning - data input to the system with a correct answers. After millions of combination AI can make a prediction base on previously input data
Common uses
- Email spam detection
- Face recognition
- Speech-to-text
- Medical image analysis
- Predicting house prices
- Credit card fraud detection
- Product recommendations
Main algorithms:
- Linear regression
- Logistic regression
- Decision tree
- Support Vector machines (SVM)
- Neural network
2) Unsupervised learning is a type of machine learning where the AI is given data without any correct answers (labels). Its job is to discover patterns or structure on its own.
Common uses
- Customer segmentation for marketing
- Finding unusual behavior (anomaly detection)
- Grouping similar documents or news articles
- Organizing photos by similar faces or scenes
- Reducing the number of variables while keeping the most important information (dimensionality reduction)
Main algorithms:
- K-Means clustering
- Hierarchical clustering
- Principal Component Analysis (PSA)
3) Semi-supervised learning mix of first and second. As an inpur sprovided small amount of labeled data and large amount of unlabeled data
4) Reinforcement learning - method of tries and errors. Agent interact with environment and receive reword or penalty depending on results. Agent is trying to maximize reward Common uses
- Playing games (Chess, Go, video games)
- Controlling robots
- some autonomous driving tasks
- Warehouse automation
- Optimizing industrial processes
- Resource allocation and scheduling
Main algorithms:
- Q-Learning
- Deep Q-Learning
- Actor critic method