Try our new curated archive search.

The Science Of AI: Appendix A - Taxonomy Of ML Statistical Algorithms

These algorithms fall under the general heading of statistical analysis.

This is an Appendix to our series of articles on Broadcast Standards – The Science Of AI.

Regression

Statistical regression algorithms suitable for use with supervised training methods:

  • Polynomial Regression.
  • Ridge Regression.
  • Lasso Regression.
  • Elastic Net Regression.
  • Support Vector Regression (SVR).
  • Decision Tree Regression.
  • Random Forest Regression.
  • Gradient Boosting Regression.
  • eXtreme Gradient Boosting (XGBoost).
  • Light Gradient Boosting Machine (LightGBM).
  • Gradient boosting on decision trees (CatBoost).
  • Bayesian Regression.
  • K-Nearest Neighbors (KNN) Regression.
Classification

Statistical classification algorithms for grouping data items into sets with supervised training methods:

  • Logistic Regression.
  • Support Vector Machines (SVM).
  • K-Nearest Neighbors (KNN) Classification.
  • Decision Trees.
  • Random Forest Classification.
  • Gradient Boosting Machines (GBM).
  • AdaBoost.
  • eXtreme Gradient Boosting (XGBoost).
  • Light Gradient Boosting Machine (LightGBM).
  • Gradient boosting on decision trees (CatBoost).
  • Naive Bayes.
  • Neural Networks.
  • Multilayer Perceptron (MLP).
  • Quadratic Discriminant Analysis (QDA).
  • Linear Discriminant Analysis (LDA).
Clustering

Clustering algorithms for use with unsupervised training methods:

  • K-Means.
  • Hierarchical Clustering.
  • Density-Based Spatial Clustering of Apps with Noise (DBSCAN).
  • Mean Shift.
  • Gaussian Mixture Models (GMM).
  • Balanced Iterative Reducing & Clustering with Hierarchies (BIRCH).
  • Affinity Propagation.
Dimensional Reduction

Dimensionality reduction algorithms for use with unsupervised training methods:

  • Principal Component Analysis (PCA).
  • Independent Component Analysis (ICA).
  • t-Distributed Stochastic Neighbor Embedding (t-SNE).
  • Linear Discriminant Analysis (LDA).
  • Factor Analysis.
  • Non-Negative Matrix Factorization (NMF).
  • UMAP (Uniform Manifold Approximation and Projection).

You might also like...

Network Traffic Engineering: Why MPEG-TS Is Still The Standard

MPEG transport stream (MPEG TS) was designed in the 1990s to deliver continuous video and audio over unreliable, one-way networks, such as satellite, terrestrial RF, and cable, where packet loss and corruption are expected. But it is still prevalent in…

Standards: Video - High Efficiency Video Coding (HEVC)

Designed to halve the bitrate of AVC while supporting resolutions up to 16K, HEVC represents a significant leap in video coding efficiency. This guide explores its profiles, tiers and levels, and examines whether it can overcome the challenges of entrenched…

SMPTE Education Launches Summer 2026 Lineup Of IP And ST 2110 Courses

Boasting two standalone courses, an intensive boot camp, and a hands-on practical lab, SMPTE Education has launched its summer 2026 Lineup of IP and ST 2110 Courses.

Standards: Video - Advanced Video Coding (AVC)

AVC remains one of the most widely deployed video codecs in the world, but navigating its profiles, levels and signaling mechanisms is far from straightforward.

Network Traffic Engineering: RIST & SRT - The Success Of ARQ Based Protocols

IP networks are inherently unreliable. We kick off this series on IP Network Traffic Engineering with a look at how RIST and SRT give broadcast engineers user-configurable control over the latency-versus-reliability trade-off for real-time media streaming.