@inproceedings{Goernitz.2011.nips,
author = {Görnitz, N and Widmer, C and Zeller, G and Kahles, A and Sonnenburg, S and Rätsch, G},
title = {{Hierarchical multitask structured output learning for large-scale sequence segmentation}},
booktitle = {Advances in Neural Information Processing Systems 24},
editor = {J. Shawe-Taylor and R.S. Zemel and P. Bartlett and F.C.N. Pereira and K.Q. Weinberger},
pages = {2690--2698},
year = {2011},
url = {http://books.nips.cc/nips24.html},
abstract = {{We present a novel regularization-based Multitask Learning (MTL) formulation for Structured Output (SO) prediction for the case of hierarchical task relations. Structured output learning often results in difficult inference problems and requires large amounts of training data to obtain accurate models. We propose to use MTL to exploit information available for related structured output learning tasks by means of hierarchical regularization. Due to the combination of example sets, the cost of training models for structured output prediction can easily become infeasible for real world applications. We thus propose an efficient algorithm based on bundle methods to solve the optimization problems resulting from MTL structured output learning. We demonstrate the performance of our approach on gene ﬁnding problems from the application domain of computational biology. We show that 1) our proposed solver achieves much faster convergence than previous methods and 2) that the Hierarchical SO-MTL approach clearly outperforms considered non-MTL methods.}}
}