🤖 Machine Learning
If you spot any errors, feel free to drop me a note at [email protected]. Machine Learning (ML) is the core engine of modern AI systems.This page is your structured gateway to every major ML paradigm, from beginner-friendly algorithms to specialized and cutting-edge methods.
📌 What You’ll Explore
Supervised Learning
Learn models trained on labeled data — including regression, classification, and sequence prediction.
Unsupervised Learning
Discover clustering, dimensionality reduction, and anomaly detection.
Semi-Supervised Learning
Techniques that leverage small labeled datasets with large unlabeled ones.
Self-Supervised Learning
Learn how modern foundation models pre-train using the data itself as supervision.
Reinforcement Learning (RL)
Agents that learn via trial-and-error, rewards, and penalties.
Online Learning
Train models incrementally as new data arrives.
Ensemble Learning
Boost accuracy with methods like bagging, boosting, and stacking.
Evolutionary Algorithms
Optimization inspired by natural selection, genetic algorithms, and swarm intelligence.
Probabilistic Models
Bayesian networks, Markov models, and statistical inference.
Graph Machine Learning
Learn algorithms for graph-structured data, including GNNs and link prediction.
Meta-Learning
“Learning to learn” — models that adapt quickly to new tasks.
🗂 ML Types Overview
| Category | Description | Example Algorithms |
|---|---|---|
| Supervised | Learns from labeled datasets. | Linear Regression, SVM, Random Forest, Neural Networks |
| Unsupervised | Finds hidden patterns in unlabeled data. | K-Means, DBSCAN, PCA, t-SNE |
| Semi-Supervised | Mix of labeled and unlabeled data. | Semi-Supervised SVM, Ladder Networks |
| Self-Supervised | Generates labels from data itself. | BERT, SimCLR, BYOL |
| Reinforcement | Learns via interaction & rewards. | Q-Learning, PPO, DQN |
| Online Learning | Updates model with incoming data. | Stochastic Gradient Descent, Passive-Aggressive |
| Ensemble | Combines multiple models. | Bagging, AdaBoost, XGBoost |
| Evolutionary | Nature-inspired optimization. | Genetic Algorithms, Particle Swarm Optimization |
| Probabilistic | Models uncertainty and inference. | Bayesian Networks, HMM |
| Graph ML | Works with graph-structured data. | GCN, GraphSAGE |
| Meta-Learning | Few-shot, adaptable learning. | MAML, Reptile |
📚 Where to Go Next
Built by Mohammad Jafrin Hossain — part of the AI Pathway series.