Multi-Task Feature Learning
Andreas Argyriou, Theodoros Evgeniou, Massimiliano Pontil · The MIT Press eBooks · 2007
We present a method for learning a low-dimensional representation which is shared across a set of multiple related tasks. The method builds upon the well-known 1-norm regularization problem using a new regularizer which controls the number of learned features common for all the tasks. We show that this problem is equivalent to a convex optimization problem and develop an iterative algorithm for solving it. The algorithm has a simple interpretation: it alternately performs a supervised and an unsupervised step, where in the latter step we learn common-across-tasks representations and in the former step we learn task-specific functions using these representations. We report experiments on a simulated and a real data set which demonstrate that the proposed method dramatically improves the per-formance relative to learning each task independently. Our algorithm can also be used, as a special case, to simply select – not learn – a few common features across the tasks.
Cited by 1,333 works. doi:10.7551/mitpress/7503.003.0010
Authors
References
- The Elements of Statistical Learning: Data Mining, Inference, and Prediction
- The Elements of Statistical Learning: Data Mining, Inference, and Prediction
- A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data
- Sharing features: efficient boosting procedures for multiclass object detection
- Maximum-Margin Matrix Factorization
- A rank minimization heuristic with application to minimum order system approximation
- For most large underdetermined systems of equations, the minimal 𝓁 1 ‐norm near‐solution approximates the sparsest near‐solution
- A Model of Inductive Bias Learning
- Exploiting Task Relatedness for Multiple Task Learning
- Learning Multiple Tasks with Kernel Methods