Collaborating differently on different topics: a multi-relational approach to multi-task learning

Sunil Kumar Gupta, Santu Rana, Dinh Phung, Svetha Venkatesh

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

1 Citation (Scopus)


Multi-task learning offers a way to benefit from synergy of multiple related prediction tasks via their joint modeling. Current multitask techniques model related tasks jointly, assuming that the tasks share the same relationship across features uniformly. This assumption is seldom true as tasks may be related across some features but not others. Addressing this problem, we propose a new multi-task learning model that learns separate task relationships along different features. This added flexibility allows our model to have a finer and differential level of control in joint modeling of tasks along different features. We formulate the model as an optimization problem and provide an efficient, iterative solution. We illustrate the behavior of the proposed model using a synthetic dataset where we induce varied feature-dependent task relationships: positive relationship, negative relationship, no relationship. Using four real datasets, we evaluate the effectiveness of the proposed model for many multi-task regression and classification problems, and demonstrate its superiority over other state-of-the-art multi-task learning models.

Original languageEnglish
Title of host publicationAdvances in Knowledge Discovery and Data Mining
Subtitle of host publication19th Pacific-Asia Conference, PAKDD 2015 Ho Chi Minh City, Vietnam, May 19–22, 2015 Proceedings, Part I
EditorsTru Cao, Ee-Peng Lim, Zhi-Hua Zhou, Tu-Bao Ho, David Cheung, Hiroshi Motoda
Place of PublicationCham Switzerland
Number of pages14
ISBN (Electronic)9783319180380
ISBN (Print)9783319180373
Publication statusPublished - 2015
Externally publishedYes
EventPacific-Asia Conference on Knowledge Discovery and Data Mining 2015 - Ho Chi Minh City, Vietnam
Duration: 19 May 201522 May 2015
Conference number: 19th (Proceedings)

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


ConferencePacific-Asia Conference on Knowledge Discovery and Data Mining 2015
Abbreviated titlePAKDD 2015
CityHo Chi Minh City
Internet address

Cite this