> ML_ALGORITHM // SUPPORT-VECTOR-MACHINES_v1.0
Support Vector Machines (SVM)
Finds the optimal separating hyperplane that maximizes the geometric margin between data classes using convex quadratic programming.
Kernel Methodsclassical-supervisedblack-boxsmall (<1k)
Back to All AlgorithmsComputational Complexity
Training Complexity:O(n^2) to O(n^3) quadratic scaling
Inference Complexity:O(n_support_vectors * p)
Hardware Profile
CPU Friendly:Yes
Requires GPU:No
Memory Footprint:high
Interpretability & Data
Interpretability Tier:black-box
Training Data Needs:small (<1k)
Interpretability Assessment
Non-linear kernels (RBF, Polynomial) operate in infinite-dimensional space; weights cannot be directly inspected.
Suitable Tasks & Supported Modalities
Suitable Tasks:
binary classificationmulticlass classificationregressionanomaly detection
Supported Modalities:
tabulartext
Implementing Libraries
Foundational Literature
Support-Vector NetworksCorinna Cortes, Vladimir N. Vapnik (1995) · Machine Learning
Common Pitfalls & Warnings
- Attempting to train on >100,000 samples leading to cubic compute stall
- Failing to standardize features leading to dominant scale distortions