Abstract:The multi-task learning support vector machines (SVMs) have recently attracted considerable attention since the conventional single task learning ones usually ignore the relatedness among multiple related tasks and train them separately. Different from the single task learning, the multi-task learning methods can capture the correlation among tasks and achieve an improved performance by training all tasks simultaneously. In this paper, we make two assumptions on the relatedness among tasks. One is that the normal vectors of the related tasks share a certain common parameter value; the other is that the models of the related tasks are close enough and share a common model. Under these assumptions, we propose two multi-task learning methods, named as MTL-aLS-SVM I and MTL-aLS-SVM II respectively, for binary classification by taking full advantages of multi-task learning and the asymmetric least squared loss. MTL-aLS-SVM I seeks for a trade-off between the maximal expectile distance for each task model and the closeness of each task model to the averaged model. MTL-aLS-SVM II can use different kernel functions for different tasks, and it is an extension of the MTL-aLS-SVM I. Both of them can be easily implemented by solving quadratic programming. In addition, we develop their special cases which include L2-SVM based multi-task learning methods (MTL-L2-SVM I and MTL-L2-SVM II) and the least squares SVM (LS-SVM) based multi-task learning methods (MTL-LS-SVM I and MTL-LS-SVM II). Although the MTL-L2-SVM II and MTL-LS-SVM II appear in the form of special cases, they are firstly proposed in this paper. The experimental results show that the proposed methods are very encouraging.

The Als-Svm Based Multi-Task Learning Classifiers

A Multi-Class Large Margin Classifier

Multi-task least-squares support vector machines

Multi-task Least Squares Twin Support Vector Machine for Classification

Low-Rank Multitask Learning based on Tensorized SVMs and LSSVMs

Multi-task Support Vector Machine Classifier with Generalized Huber Loss

Multitask Multiclass Support Vector Machines

Efficacy of Regularized Multitask Learning Based on SVM Models.

Multitask twin support vector machines

Multitask Multiclass Support Vector Machines: Model and Experiments

Least-Squares Support Vector Machine for Semi-Supervised Multi-Tasking

A Rough Margin-Based Multi-Task Ν-Twin Support Vector Machine for Pattern Classification.

Multi-task Nonparallel Support Vector Machine for Classification

Multi-task Support Vector Machine with Pinball Loss.

Multi-Task Clustering with Model Relation Learning.

Tensorized LSSVMS for Multitask Regression

A novel multi-task twin-hypersphere support vector machine for classification

Multi-Task Learning with Group-Specific Feature Space Sharing

Least squares recursive projection twin support vector machine for multi-class classification

On Better Exploring and Exploiting Task Relationships in Multitask Learning: Joint Model and Feature Learning.

ML2S-SVM: multi-label least-squares support vector machine classifiers.