Jeffreys prior regularization for logistic regression
| Autor: | Raviv Raich, Tam Nguyen, Phung Lai |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
0301 basic medicine
Regularization perspectives on support vector machines Logistic regression Regularization (mathematics) Tikhonov regularization 03 medical and health sciences symbols.namesake 030104 developmental biology Statistics symbols Maximum a posteriori estimation Applied mathematics Fisher information Multinomial logistic regression Mathematics Jeffreys prior |
| Zdroj: | SSP |
| Popis: | Logistic regression is a statistical model widely used for solving classification problems. Maximum likelihood is used train the model parameters. When data from two classes is linearly separable, maximum likelihood is ill-posed. To address this problem as well as to handle over-fitting issues, regularization is commonly considered. A regularization coefficient is used to control the tradeoff between model complexity and data fit and cross-validation is applied to determine the coefficient. In this paper, we develop a regularization framework for logistic regression using Jeffreys prior, which is free of any tuning parameters. Our experiments show that the proposed regularization outperforms other well-known regularization approaches. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|