Blockwise Coordinate Descent Procedures for the Multi-task Lasso, with Applications to Neural Semantic Basis Discovery

Han Liu*, Mark Palatucci, Jian Zhang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Scopus citations

Abstract

We develop a cyclical blockwise coordinate descent algorithm for the multi-task Lasso that effciently solves problems with thousands of features and tasks. The main result shows that a closed-form Winsorization operator can be obtained for the sup-norm penalized least squares regression. This allows the algorithm to nd solutions to very largescale problems far more eficiently than existing methods. This result complements the pioneering work of Friedman, et al. (2007) for the single-task Lasso. As a case study, we use the multi-task Lasso as a variable selector to discover a semantic basis for predicting human neural activation. The learned solution outperforms the standard basis for this task on the majority of test participants, while requiring far fewer assumptions about cognitive neuroscience. We demonstrate how this learned basis can yield insights into how the brain represents the meanings of words.

Original languageEnglish (US)
Title of host publicationProceedings of the 26th Annual International Conference on Machine Learning, ICML'09
DOIs
StatePublished - 2009
Event26th Annual International Conference on Machine Learning, ICML'09 - Montreal, QC, Canada
Duration: Jun 14 2009Jun 18 2009

Publication series

NameACM International Conference Proceeding Series
Volume382

Other

Other26th Annual International Conference on Machine Learning, ICML'09
CountryCanada
CityMontreal, QC
Period6/14/096/18/09

ASJC Scopus subject areas

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Blockwise Coordinate Descent Procedures for the Multi-task Lasso, with Applications to Neural Semantic Basis Discovery'. Together they form a unique fingerprint.

Cite this