File(s) under permanent embargo
Collaborating differently on different topics: a multi-relational approach to multi-task learning
chapter
posted on 2015-01-01, 00:00 authored by Sunil GuptaSunil Gupta, Santu RanaSantu Rana, Quoc-Dinh Phung, Svetha VenkateshSvetha VenkateshMulti-task learning offers a way to benefit from synergy of multiple related prediction tasks via their joint modeling. Current multi-task 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
History
Event
Pacific-Asia Conference on Knowledge Discovery and Data MiningTitle of book
Advances in knowledge discovery and data mining 19th Pacific-Asia Conference, PAKDD 2015, Ho Chi Minh City, Vietnam, May 19-22, 2015, Proceedings, Part IVolume
9077Series
Lecture notes in computer science; v.9077Chapter number
24Pagination
303 - 316Publisher
SpringerLocation
VietnamPlace of publication
Berlin, GermanyPublisher DOI
Start date
2015-01-01End date
2015-01-01ISSN
0302-9743eISSN
1611-3349ISBN-13
9783319180373Language
engPublication classification
B Book chapter; B1 Book chapterCopyright notice
2015, SpringerExtent
58Editor/Contributor(s)
T Cao, E Lim, Z Zhou, T Ho, D Cheung, H MotodaUsage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC