Multi-Task Averaging
By Sergey Feldman, Maya R. Gupta, Andrew Béla Frigyik. 2011
16 citations recorded by the source corpus.
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 averages. We derive the optimal amount of regularization, and show that it can be effectively estimated. Simulations and real data experiments demonstrate that MTA outperforms both maximum likelihood and James-Stein estimators, and that our approach to estimating the amount of regularization rivals cross-validation in performance but is more computationally efficient.
Subfields: Artificial Intelligence, Computational Mechanics
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Kenji Fukumizu
- Bernhard Schölkopf
- Bharath K. Sriperumbudur
- Krikamol Muandet
- Arthur Gretton
- David Lopez-Paz
- Dino Sejdinovic
- Le Song
In Computational Mechanics
- Bernhard Schölkopf
- Krikamol Muandet
- Arthur Gretton
- Kenji Fukumizu
- Andrew Béla Frigyik
- Bharath K. Sriperumbudur
- Sergey Feldman
- Maya R. Gupta
Citations in the corpus
- Kernel Mean Shrinkage Estimators (2014, background): “mean embeddings, it is natural to ask if an improvement can be gained by estimating them jointly. On the one hand, one may view this setting more generally as a multitask learning recently studied by Feldman et al. (2012) and MartnezRego and Pontil (2013). On the other hand, it is often assumed a pr”
- Revisiting Stein's paradox: multi-task averaging (2014, background): “A short version of this paper was published in NIPS 2012 (Feldman et al., 2012).”