TY - GEN
T1 - Multi-stage multi-task feature learning
AU - Gong, Pinghua
AU - Ye, Jieping
AU - Zhang, Changshui
PY - 2012/12/1
Y1 - 2012/12/1
N2 - Multi-task sparse feature learning aims to improve the generalization performance by exploiting the shared features among tasks. It has been successfully applied to many applications including computer vision and biomedical informatics. Most of the existing multi-task sparse feature learning algorithms are formulated as a convex sparse regularization problem, which is usually suboptimal, due to its looseness for approximating an l0-type regularizer. In this paper, we propose a non-convex formulation for multi-task sparse feature learning based on a novel regularizer. To solve the non-convex optimization problem, we propose a Multi-Stage Multi-Task Feature Learning (MSMTFL) algorithm. Moreover, we present a detailed theoretical analysis showing that MSMTFL achieves a better parameter estimation error bound than the convex formulation. Empirical studies on both synthetic and real-world data sets demonstrate the effectiveness of MSMTFL in comparison with the state of the art multi-task sparse feature learning algorithms.
AB - Multi-task sparse feature learning aims to improve the generalization performance by exploiting the shared features among tasks. It has been successfully applied to many applications including computer vision and biomedical informatics. Most of the existing multi-task sparse feature learning algorithms are formulated as a convex sparse regularization problem, which is usually suboptimal, due to its looseness for approximating an l0-type regularizer. In this paper, we propose a non-convex formulation for multi-task sparse feature learning based on a novel regularizer. To solve the non-convex optimization problem, we propose a Multi-Stage Multi-Task Feature Learning (MSMTFL) algorithm. Moreover, we present a detailed theoretical analysis showing that MSMTFL achieves a better parameter estimation error bound than the convex formulation. Empirical studies on both synthetic and real-world data sets demonstrate the effectiveness of MSMTFL in comparison with the state of the art multi-task sparse feature learning algorithms.
UR - http://www.scopus.com/inward/record.url?scp=84877724400&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84877724400&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84877724400
SN - 9781627480031
T3 - Advances in Neural Information Processing Systems
SP - 1988
EP - 1996
BT - Advances in Neural Information Processing Systems 25
T2 - 26th Annual Conference on Neural Information Processing Systems 2012, NIPS 2012
Y2 - 3 December 2012 through 6 December 2012
ER -