Featured algorithm
— running liveTaxonomy constellation
Every model, mapped by family. Hover a node to light up its branch.
The catalogue
— tilt · spotlight · flip for the mathSupervised
Linear Regression
Supervised · Regression
Regression
ŷ = w · x + b
Learns a best‑fit line to predict continuous values.
Linear Regression
Logistic Regression
Supervised · Classification
Classification
p = σ(w · x + b)
Models class probability using a sigmoid/softmax decision boundary.
Logistic Regression
Support Vector Machine (SVM)
Supervised · Classification
Classification
min ½‖w‖² s.t. yᵢ(w·xᵢ+b) ≥ 1
Finds a maximum‑margin separating hyperplane (kernelizable).
Support Vector Machine (SVM)
K‑Nearest Neighbors (KNN)
Supervised · Classification/Regression
Classification/Regression
ŷ = vote{ y : x ∈ Nₖ(x) }
Predicts from the labels/values of the nearest points in feature space.
K‑Nearest Neighbors (KNN)
Decision Tree
Supervised · Tree‑based
Tree‑based
split → max information gain
Learns hierarchical if‑then splits to predict class or value.
Decision Tree
Random Forest
Supervised · Ensemble
Ensembles
ŷ = mode(T₁ … T_B)
Bagging ensemble of trees; reduces variance and improves robustness.
Random Forest
Gradient Boosting
Supervised · Ensemble
Ensembles
Fₘ = Fₘ₋₁ + ν · hₘ
Builds learners sequentially to correct errors (e.g. XGBoost/LightGBM style).
Gradient Boosting
MLP (Neural Network)
Supervised · Neural Network
Neural Networks
a⁽ˡ⁾ = σ(W a⁽ˡ⁻¹⁾ + b)
Learns non‑linear mappings using layered neurons and backpropagation.
MLP (Neural Network)
Unsupervised
K‑Means
Unsupervised · Clustering
Clustering
argmin Σ ‖x − μₖ‖²
Partitions data into k clusters by iteratively updating centroids.
K‑Means
DBSCAN
Unsupervised · Density‑based Clustering
Clustering
core: |Nε(p)| ≥ minPts
Finds dense regions and marks sparse points as noise/outliers.
DBSCAN
PCA
Unsupervised · Dimensionality Reduction
Dimensionality Reduction
max Var(Xw), ‖w‖ = 1
Projects data onto principal components maximizing variance.
PCA
t‑SNE
Unsupervised · Dimensionality Reduction
Dimensionality Reduction
min KL(P ‖ Q)
Visualizes high‑dimensional data by preserving local neighborhoods.
t‑SNE
Reinforcement
Q‑Learning
Reinforcement · Value‑Based
Value‑Based
Q ← Q + α[ r + γ·maxQ′ − Q ]
Learns state‑action values to choose actions that maximize reward.
Q‑Learning
DQN
Reinforcement · Deep RL
Deep RL
L = ( r + γ·maxQ(s′;θ⁻) − Q(s,a;θ) )²
Approximates Q‑values with a neural network for high‑dim inputs.
DQN
PPO
Reinforcement · Actor‑Critic
Actor‑Critic
max E[ min(rₜAₜ, clip(rₜ)·Aₜ) ]
Stabilizes policy updates using a clipped objective (widely used in practice).
PPO
Interactive playground
You are the query point. K-Nearest Neighbours finds your 5 closest points and votes on the class — live.
k=5 vote —