A Unifying Framework for Typical Multi-Task Multiple Kernel Learning Problems
This work provides a unifying framework for researchers in kernel-based learning, addressing the complexity of designing custom algorithms for different MKL problems, though it is incremental as it builds on existing MKL and Multi-Task Learning methods.
The authors tackled the problem of diverse Multi-Kernel Learning (MKL) formulations requiring tailored algorithms by proposing a general Multi-Task MKL framework that unifies existing approaches and enables solving them with a simple algorithm, and they demonstrated its flexibility by formulating and testing a new Partially-Shared Common Space Multi-Task MKL problem.
Over the past few years, Multi-Kernel Learning (MKL) has received significant attention among data-driven feature selection techniques in the context of kernel-based learning. MKL formulations have been devised and solved for a broad spectrum of machine learning problems, including Multi-Task Learning (MTL). Solving different MKL formulations usually involves designing algorithms that are tailored to the problem at hand, which is, typically, a non-trivial accomplishment. In this paper we present a general Multi-Task Multi-Kernel Learning (Multi-Task MKL) framework that subsumes well-known Multi-Task MKL formulations, as well as several important MKL approaches on single-task problems. We then derive a simple algorithm that can solve the unifying framework. To demonstrate the flexibility of the proposed framework, we formulate a new learning problem, namely Partially-Shared Common Space (PSCS) Multi-Task MKL, and demonstrate its merits through experimentation.