Background: The lack of sucient training data is the limiting factor for many Machine Learning applications in Computational Biology. If data is available for several dierent but related problem domains, Multitask Learning algorithms can be used to learn a model based on all available information. In Bioinformatics, many problems can be cast into the Multitask Learning scenario by incorporating data from several organisms. However, combining information from several tasks requires the careful consideration of the degree of similarity between tasks. Our proposed method simultaneously learns (or renes) the similarity between tasks along with the Multitask Learning Classier. This is done by formulating the Multitask Learning problem as Multiple Kernel Learning, using the recently published q-Norm MKL algorithm.
Results: We demonstrate the performance of our method on two problems from Computational Biology. First, we show that our method is able to improve performance on a splice-site dataset with given hierarchical task structure by rening the task relationships. Second, we consider an MHC-I dataset, for which we assume no knowledge about the degree of task relatedness. Here, we are able to learn the task similarity from scratch along with the Multitask classiers. In both cases, we clearly outperform baseline methods that we compare against.
Conclusions: We present a novel approach to Multitask Learning that learns task similarity along with the classiers. The framework is very general as it allows to incorporate prior knowledge about tasks relationships if available, but is also able to identify task similarities in absence of additional information. Both variants show promising results in applications from Computational Biology.