Adaptive and robust multi-task learning
Yaqi Duan, Kaizheng Wang
Abstract
We study the multitask learning problem that aims to simultaneously analyze multiple data sets collected from different sources and learn one model for each of them. We propose a family of adaptive methods that automatically utilize possible similarities among those tasks while carefully handling their differences. We derive sharp statistical guarantees for the methods and prove their robustness against outlier tasks. Numerical experiments on synthetic and real data sets demonstrate the efficacy of our new methods.
Topics & Concepts
MathematicsMulti-task learningTask (project management)Artificial intelligenceEconometricsMachine learningComputer scienceManagementEconomicsDomain Adaptation and Few-Shot LearningSparse and Compressive Sensing TechniquesMachine Learning and Algorithms