Multi-Task Learning Using Neighborhood Kernels
This work addresses kernel optimization for multi-task learning, offering a novel method that can also apply to single-task learning, but it appears incremental as it builds on existing kernel learning approaches.
The paper tackles the problem of learning kernels in multi-task learning by introducing a neighborhood-defining kernel framework, resulting in an algorithm that consistently outperforms traditional kernel learning methods in classification and regression tasks.
This paper introduces a new and effective algorithm for learning kernels in a Multi-Task Learning (MTL) setting. Although, we consider a MTL scenario here, our approach can be easily applied to standard single task learning, as well. As shown by our empirical results, our algorithm consistently outperforms the traditional kernel learning algorithms such as uniform combination solution, convex combinations of base kernels as well as some kernel alignment-based models, which have been proven to give promising results in the past. We present a Rademacher complexity bound based on which a new Multi-Task Multiple Kernel Learning (MT-MKL) model is derived. In particular, we propose a Support Vector Machine-regularized model in which, for each task, an optimal kernel is learned based on a neighborhood-defining kernel that is not restricted to be positive semi-definite. Comparative experimental results are showcased that underline the merits of our neighborhood-defining framework in both classification and regression problems.