Supervised
Linear Regression
Supervised · Regression
Regression
Learns a best‑fit line to predict continuous values.
Logistic Regression
Supervised · Classification
Classification
Models class probability using a sigmoid/softmax decision boundary.
Support Vector Machine (SVM)
Supervised · Classification
Classification
Finds a maximum‑margin separating hyperplane (kernelizable).
K‑Nearest Neighbors (KNN)
Supervised · Classification/Regression
Classification/Regression
Predicts from the labels/values of the nearest points in feature space.
Decision Tree
Supervised · Tree‑based
Tree‑based
Learns hierarchical if‑then splits to predict class or value.
Random Forest
Supervised · Ensemble
Ensembles
Bagging ensemble of trees; reduces variance and improves robustness.
Gradient Boosting
Supervised · Ensemble
Ensembles
Builds learners sequentially to correct errors (e.g. XGBoost/LightGBM style).
MLP (Neural Network)
Supervised · Neural Network
Neural Networks
Learns non‑linear mappings using layered neurons and backpropagation.
Unsupervised
K‑Means
Unsupervised · Clustering
Clustering
Partitions data into k clusters by iteratively updating centroids.
DBSCAN
Unsupervised · Density‑based Clustering
Clustering
Finds dense regions and marks sparse points as noise/outliers.
PCA
Unsupervised · Dimensionality Reduction
Dimensionality Reduction
Projects data onto principal components maximizing variance.
t‑SNE
Unsupervised · Dimensionality Reduction
Dimensionality Reduction
Visualizes high‑dimensional data by preserving local neighborhoods.
Reinforcement
Q‑Learning
Reinforcement · Value‑Based
Value‑Based
Learns state‑action values to choose actions that maximize reward.
DQN
Reinforcement · Deep RL
Deep RL
Approximates Q‑values with a neural network for high‑dim inputs.
PPO
Reinforcement · Actor‑Critic
Actor‑Critic
Stabilizes policy updates using a clipped objective (widely used in practice).