Sparse coding for multitask and transfer learning

Abstract

We investigate the use of sparse coding and dictionary learning in the context of multitask and transfer learning. The central assumption of our learning method is that the tasks parameters are well approximated by sparse linear combinations of the atoms of a dictionary on a high or infinite dimensional space. This assumption, together with the large quantity of available data in the multitask and transfer learning settings, allows a principled choice of the dictionary. We provide bounds on the generalization error of this approach, for both settings. Numerical experiments on one synthetic and two real datasets show the advantage of our method over single task learning, a previous method based on orthogonal and dense representation of the tasks and a related method learning task grouping.

International Conference on Machine Learning 2013

References in corpus (2)

• Theoretical Models of Learning to Learn
• Proximal Methods for Hierarchical Sparse Coding

Cited by in corpus (5)

• On Better Exploring and Exploiting Task Relationships in Multi-Task Learning: Joint Model and Feature Learning
• A Clustering Approach to Learn Sparsely-Used Overcomplete Dictionaries
• On The Identifiability of Mixture Models from Grouped Samples
• Interleaved Multitask Learning with Energy Modulated Learning Progress
• Deep Asymmetric Multi-task Feature Learning