TY - GEN
T1 - Linking heterogeneous input spaces with pivots for multi-task learning
AU - He, Jingrui
AU - Liu, Yan
AU - Yang, Qiang
N1 - Publisher Copyright:
Copyright © SIAM.
PY - 2014
Y1 - 2014
N2 - Most existing works on multi-task learning (MTL) assume the same input space for different tasks. In this paper, we address a general setting where different tasks have heterogeneous input spaces. This setting has a lot of potential applications, yet it poses new algorithmic challenges - how can we link seemingly uncorrelated tasks to mutually boost their learning performance? Our key observation is that in many real applications, there might exist some correspondence among the inputs of different tasks, which is referred to as pivots. For such applications, we first propose a learning scheme for multiple tasks and analyze its generalization performance. Then we focus on the problems where only a limited number of the pivots are available, and propose a general framework to leverage the pivot information. The idea is to map the heterogeneous input spaces to a common space, and construct a single prediction model in this space for all the tasks. We further propose an effective optimization algorithm to find both the mappings and the prediction model. Experimental results demonstrate its effectiveness, especially with very limited number of pivots.
AB - Most existing works on multi-task learning (MTL) assume the same input space for different tasks. In this paper, we address a general setting where different tasks have heterogeneous input spaces. This setting has a lot of potential applications, yet it poses new algorithmic challenges - how can we link seemingly uncorrelated tasks to mutually boost their learning performance? Our key observation is that in many real applications, there might exist some correspondence among the inputs of different tasks, which is referred to as pivots. For such applications, we first propose a learning scheme for multiple tasks and analyze its generalization performance. Then we focus on the problems where only a limited number of the pivots are available, and propose a general framework to leverage the pivot information. The idea is to map the heterogeneous input spaces to a common space, and construct a single prediction model in this space for all the tasks. We further propose an effective optimization algorithm to find both the mappings and the prediction model. Experimental results demonstrate its effectiveness, especially with very limited number of pivots.
UR - http://www.scopus.com/inward/record.url?scp=84959910535&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84959910535&partnerID=8YFLogxK
U2 - 10.1137/1.9781611973440.21
DO - 10.1137/1.9781611973440.21
M3 - Conference contribution
AN - SCOPUS:84959910535
T3 - SIAM International Conference on Data Mining 2014, SDM 2014
SP - 181
EP - 189
BT - SIAM International Conference on Data Mining 2014, SDM 2014
A2 - Zaki, Mohammed J.
A2 - Banerjee, Arindam
A2 - Parthasarathy, Srinivasan
A2 - Ning-Tan, Pang
A2 - Obradovic, Zoran
A2 - Kamath, Chandrika
PB - Society for Industrial and Applied Mathematics Publications
T2 - 14th SIAM International Conference on Data Mining, SDM 2014
Y2 - 24 April 2014 through 26 April 2014
ER -