> ML_ALGORITHM // DISCRIMINANT-ANALYSIS_v1.0
Discriminant Analysis (LDA / QDA)
Generative classification technique that models class densities as multivariate Gaussians, producing optimal linear or quadratic decision boundaries.
Linear & Generalized Modelsclassical-supervisedhigh-intrinsicsmall (<1k)
Back to All AlgorithmsComputational Complexity
Training Complexity:O(n * p^2 + p^3) matrix inversion
Inference Complexity:O(p)
Hardware Profile
CPU Friendly:Yes
Requires GPU:No
Memory Footprint:low
Interpretability & Data
Interpretability Tier:high-intrinsic
Training Data Needs:small (<1k)
Interpretability Assessment
Linear discriminant projections provide clear orthogonal class separation axes.
Suitable Tasks & Supported Modalities
Suitable Tasks:
multiclass classificationdimensionality reduction
Supported Modalities:
tabular
Implementing Libraries
Foundational Literature
The Use of Multiple Measurements in Taxonomic ProblemsRonald A. Fisher (1936) · Annals of Eugenics
Common Pitfalls & Warnings
- Singular covariance matrix when number of features exceeds sample count (p > n)
- Sensitivity to heavy-tailed non-Gaussian distributions