Inductive multi-task learning with multiple view data
Jintao Zhang, Jun Huan · ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD) · 2012
In many real-world applications, it is becoming common to have data extracted from multiple diverse sources, known as "multi-view" data. Multi-view learning (MVL) has been widely studied in many applications, but existing MVL methods learn a single task individually. In this paper, we study a new direction of multi-view learning where there are multiple related tasks with multi-view data (i.e. multi-view multi-task learning, or MVMT Learning). In our MVMT learning methods, we learn a linear mapping for each view in each task. In a single task, we use co-regularization to obtain functions that are in-agreement with each other on the unlabeled samples and achieve low classification errors on the labeled samples simultaneously. Cross different tasks, additional regularization functions are utilized to ensure the functions that we learn in each view are similar. We also developed two extensions of the MVMT learning algorithm. One extension handles missing views and the other handles non-uniformly related tasks. Experimental studies on three real-world data sets demonstrate that our MVMT methods significantly outperform the existing state-of-the-art methods.
Cited by 130 works. doi:10.1145/2339530.2339617
Authors
References
- Combining labeled and unlabeled data with co-training
- A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data
- Multi-Task Feature Learning
- Learning multi-label scene classification
- Regularized multi--task learning
- Analyzing the effectiveness and applicability of co-training
- NUS-WIDE
- Automating the Construction of Internet Portals with Machine Learning
- PAC Generalization Bounds for Co-training
- A Co-Regularization Approach to Semi-supervised Learning with Multiple Views