A Convex Formulation for Learning Task Relationships in Multi-Task Learning
Abstract:Multi-task learning is a learning paradigm which seeks to improve the generalization performance of a learning task with the help of some other related tasks. In this paper, we propose a regularization formulation for learning the relationships between tasks in multi-task learning. This formulation can be viewed as a novel generalization of the regularization framework for single-task learning. Besides modeling positive task correlation, our method, called multi-task relationship learning (MTRL), can also describe negative task correlation and identify outlier tasks based on the same underlying principle. Under this regularization framework, the objective function of MTRL is convex. For efficiency, we use an alternating method to learn the optimal model parameters for each task as well as the relationships between tasks. We study MTRL in the symmetric multi-task learning setting and then generalize it to the asymmetric setting as well. We also study the relationships between MTRL and some existing multi-task learning methods. Experiments conducted on a toy problem as well as several benchmark data sets demonstrate the effectiveness of MTRL.
暂无分享,去 创建一个
[1] Sebastian Thrun,et al. Is Learning The n-th Thing Any Easier Than Learning The First? , 1995, NIPS.
[2] Sebastian Thrun,et al. Discovering Structure in Multiple Learning Tasks: The TC Algorithm , 1996, ICML.
[3] Rich Caruana,et al. Multitask Learning , 1997, Machine-mediated learning.
[4] Stephen P. Boyd,et al. Applications of second-order cone programming , 1998 .
[5] A. Rukhin. Matrix Variate Distributions , 1999, The Multivariate Normal Distribution.
[6] Tom Heskes,et al. Task Clustering and Gating for Bayesian Multitask Learning , 2003, J. Mach. Learn. Res..
[7] S. Keerthi,et al. SMO algorithm for least squares SVM , 2003, Proceedings of the International Joint Conference on Neural Networks, 2003..
[8] Jonathan Baxter,et al. A Bayesian/Information Theoretic Model of Learning to Learn via Multiple Task Sampling , 1997, Machine Learning.
[9] Johan A. K. Suykens,et al. Benchmarking Least Squares Support Vector Machine Classifiers , 2004, Machine Learning.
[10] Massimiliano Pontil,et al. Regularized multi--task learning , 2004, KDD.
[11] Charles A. Micchelli,et al. Learning Multiple Tasks with Kernel Methods , 2005, J. Mach. Learn. Res..
[12] Mikhail Belkin,et al. Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples , 2006, J. Mach. Learn. Res..
[13] Rajat Raina,et al. Constructing informative priors using transfer learning , 2006, ICML.
[14] Stephen P. Boyd,et al. Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.
[15] Christopher M. Bishop,et al. Pattern Recognition and Machine Learning (Information Science and Statistics) , 2006 .
[16] Kumar Chellapilla,et al. Personalized handwriting recognition via biased regularization , 2006, ICML.
[17] Edwin V. Bonilla,et al. Multi-task Gaussian Process Prediction , 2007, NIPS.
[18] Radford M. Neal. Pattern Recognition and Machine Learning , 2007, Technometrics.
[19] Charles A. Micchelli,et al. A Spectral Regularization Framework for Multi-Task Structure Learning , 2007, NIPS.
[20] Masashi Sugiyama,et al. Multi-Task Learning via Conic Programming , 2007, NIPS.
[21] Lawrence Carin,et al. Multi-Task Learning for Classification with Dirichlet Process Priors , 2007, J. Mach. Learn. Res..
[22] Massimiliano Pontil,et al. Convex multi-task feature learning , 2008, Machine Learning.
[23] Jean-Philippe Vert,et al. Clustered Multi-Task Learning: A Convex Formulation , 2008, NIPS.
[24] Eric Eaton,et al. Modeling Transfer Relationships Between Learning Tasks for Improved Inductive Transfer , 2008, ECML/PKDD.
[25] Hal Daumé,et al. Bayesian Multitask Learning with Latent Hierarchies , 2009, UAI.
[26] Jieping Ye,et al. Multi-Task Feature Learning Via Efficient l2, 1-Norm Minimization , 2009, UAI.
[27] Qiang Yang,et al. A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.
[28] Dit-Yan Yeung,et al. Multi-Task Learning using Generalized t Process , 2010, AISTATS.