Author: Kshirsagar, Meghana; Carbonell, Jaime; Klein-Seetharaman, Judith
Title: Multitask learning for host–pathogen protein interactions Document date: 2013_7_1
ID: sdgt2ms5_64
Snippet: Mean Multi-task Learning (Mean MTL): This is a logistic regression-based implementation of the multitask SVM model proposed by (Evgeniou and Pontil, 2004) . The important feature of this work is the use of a regularizer that penalizes the difference between a model and the 'mean' model formed by averaging over models from all m tasks. In the original paper, the loss functions lðw i Þ were all hinge loss. Because we find that logistic regression.....
Document: Mean Multi-task Learning (Mean MTL): This is a logistic regression-based implementation of the multitask SVM model proposed by (Evgeniou and Pontil, 2004) . The important feature of this work is the use of a regularizer that penalizes the difference between a model and the 'mean' model formed by averaging over models from all m tasks. In the original paper, the loss functions lðw i Þ were all hinge loss. Because we find that logistic regression does better on our datasets, we replaced the original hinge loss function by logistic loss. The objective we use is shown in Equation (10).
Search related documents:
Co phrase search for related documents- hinge loss and loss function: 1, 2
- hinge loss function and loss function: 1, 2
- important feature and logistic regression: 1, 2, 3, 4
- important feature and loss function: 1, 2
- important feature and SVM model: 1
- logistic loss and loss function: 1, 2
- logistic regression and loss function: 1, 2, 3, 4, 5, 6, 7, 8, 9
- logistic regression and Multi task: 1, 2, 3
- logistic regression and SVM model: 1, 2, 3
- loss function and Multi task: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10
- loss function and Multi task learning: 1, 2, 3, 4
- loss function and original paper: 1
Co phrase search for related documents, hyperlinks ordered by date