方法证据记录
Regularized Support Vector Machine
Regularized Support Vector Machine extends the classic SVM by explicitly controlling the trade-off between margin maximization and training error through an L1 or L2 penalty parameter. The soft-margin formulation introduced by Cortes and Vapnik in 1995 is itself a regularized model, and later L1-SVM variants additionally promote feature sparsity, enabling automatic variable selection in high-dimensional settings.
源记录
引文逐字复制自方法源记录。这些引文不代表任何层级的验证。
Regularized Support Vector Machine (L1/L2-penalized SVM)
分类方法记录 · ml-model / machine-learning
- Cortes, C. & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273–297. · DOI 10.1007/BF00994018
- Zhu, J., Rosset, S., Tibshirani, R. & Hastie, T. (2004). 1-norm support vector machines. Advances in Neural Information Processing Systems (NIPS), 16. · URL
精选声明
声明已持久化到证据分类账中，每个声明都有自己的评估。
尚无精选声明
当分类账中没有声明时，此视图不会自行创建声明评估。
相关方法
从方法图中生成，显示为机器建议的关系 — 不推断任何证据声明。