Multi-Task Averaging

2011/07/21 by Sergey Feldman, Feldman, Sergey, Bela A. Frigyik +3
Mathematics · #FOS: Computer and information sciences #Machine Learning (stat.ML) #Methodology (stat.ME) #stat.ME #stat.ML

paper · pdf · doi:10.48550/arxiv.1107.4390

totally redone paper

arxiv created 2012/08/24 · arxiv updated 2015/03/19

Abstract

We present a multi-task learning approach to jointly estimate the means of multiple independent data sets. The proposed multi-task averaging (MTA) algorithm results in a convex combination of the single-task maximum likelihood estimates. We derive the optimal minimum risk estimator and the minimax estimator, and show that these estimators can be efficiently estimated. Simulations and real data experiments demonstrate that MTA estimators often outperform both single-task and James-Stein estimators.

Related

• High-Dimensional Multi-Task Averaging and Application to Kernel Mean Embedding
• Adaptive and Robust Multi-Task Learning
• High Dimensional Data Enrichment: Interpretable, Fast, and Data-Efficient
• Analysing Multi-Task Regression via Random Matrix Theory with Application to Time Series Forecasting
• Multi-task learning on the edge: cost-efficiency and theoretical optimality