About: Multi-task learning

Facets (new session)
Description
Metadata
Settings
- Rule:
- Inverse Functional Properties:
- "Same As":

About: Multi-task learning Goto Sponge NotDistinct Permalink

An Entity of Type : dbo:ProgrammingLanguage, within Data Space : dbpedia.demo.openlinksw.com associated with source document(s)
QRcode icon

http://dbpedia.demo.openlinksw.com/c/4rnRwLRQc6

Multi-task learning (MTL) is a subfield of machine learning in which multiple learning tasks are solved at the same time, while exploiting commonalities and differences across tasks. This can result in improved learning efficiency and prediction accuracy for the task-specific models, when compared to training the models separately. Early versions of MTL were called "hints". In a widely cited 1997 paper, Rich Caruana gave the following characterization:

Attributes	Values
rdf:type	programming language
rdfs:label	Multi-task learning (en)
rdfs:comment	Multi-task learning (MTL) is a subfield of machine learning in which multiple learning tasks are solved at the same time, while exploiting commonalities and differences across tasks. This can result in improved learning efficiency and prediction accuracy for the task-specific models, when compared to training the models separately. Early versions of MTL were called "hints". In a widely cited 1997 paper, Rich Caruana gave the following characterization: (en)
dct:subject	Machine learning
Wikipage page ID	938663 (xsd:integer)
Wikipage revision ID	1104310342 (xsd:integer)
Link from a Wikipage to another Wikipage	Sparse array Decision trees Bergman kernel Inductive bias .NET Framework Conditional random field Convex optimization General game playing Generalization error Orthogonal Multiclass classification Convolutional neural network Machine learning Complete metric space Kernel methods for vector output Multi-label classification Machine learning C Sharp (programming language) Transfer learning Feature space Laplacian matrix Linear combination Inductive transfer Adjacency matrix Stochastic gradient descent Regularization (mathematics) Vector-valued function Artificial intelligence Artificial neural network Coercive function Regularization by spectral filtering Automated machine learning Financial modeling Human-based genetic algorithm Inner product space Loss function Robot learning Evolutionary computation Multitask optimization Overfitting Classifier (mathematics) Representation learning Spam filtering dbr:GoogLeNet
Link from a Wikipage to an external page	https://web.archive.org/web/20041118134329/http:/big.cs.uiuc.edu/webpage/cumulativeLearning/cumulativeLearning.html https://web.archive.org/web/20131224113826/http:/klcl.pku.edu.cn/member/sunxu/code.htm http://www.public.asu.edu/~jye02/Software/MALSAR/index.html http://www.cse.wustl.edu/~kilian/research/multitasklearning/multitasklearning.html
sameAs	Multi-task learning Multi-task learning Multi-task learning Multi-task learning Multi-task learning
dbp:wikiPageUsesTemplate	dbt:Div_col dbt:Div_col_end dbt:Math dbt:Mvar dbt:NumBlk dbt:Reflist dbt:Short_description dbt:EquationRef dbt:Proof dbt:EquationNote dbt:Mathcal dbt:Math_theorem
has abstract	Multi-task learning (MTL) is a subfield of machine learning in which multiple learning tasks are solved at the same time, while exploiting commonalities and differences across tasks. This can result in improved learning efficiency and prediction accuracy for the task-specific models, when compared to training the models separately. Early versions of MTL were called "hints". In a widely cited 1997 paper, Rich Caruana gave the following characterization: Multitask Learning is an approach to inductive transfer that improves generalization by using the domain information contained in the training signals of related tasks as an inductive bias. It does this by learning tasks in parallel while using a shared representation; what is learned for each task can help other tasks be learned better. In the classification context, MTL aims to improve the performance of multiple classification tasks by learning them jointly. One example is a spam-filter, which can be treated as distinct but related classification tasks across different users. To make this more concrete, consider that different people have different distributions of features which distinguish spam emails from legitimate ones, for example an English speaker may find that all emails in Russian are spam, not so for Russian speakers. Yet there is a definite commonality in this classification task across users, for example one common feature might be text related to money transfer. Solving each user's spam classification problem jointly via MTL can let the solutions inform each other and improve performance. Further examples of settings for MTL include multiclass classification and multi-label classification. Multi-task learning works because regularization induced by requiring an algorithm to perform well on a related task can be superior to regularization that prevents overfitting by penalizing all complexity uniformly. One situation where MTL may be particularly helpful is if the tasks share significant commonalities and are generally slightly under sampled. However, as discussed below, MTL has also been shown to be beneficial for learning unrelated tasks. (en)
gold:hypernym	Approach
prov:wasDerivedFrom	wikipedia-en:Multi-task_learning?oldid=1104310342&ns=0
page length (characters) of wiki page	31255 (xsd:nonNegativeInteger)
foaf:isPrimaryTopicOf	wikipedia-en:Multi-task_learning
is Link from a Wikipage to another Wikipage of	Bayesian interpretation of kernel regularization Bergman kernel Deep learning General game playing Multiclass classification Multitask learning Applications of multi-task learning Feature hashing Federated learning Kernel methods for vector output Transfer learning Matrix regularization Artificial general intelligence SpaCy MTL Multitask optimization Outline of machine learning
is Wikipage redirect of	Multitask learning Applications of multi-task learning
is Wikipage disambiguates of	MTL
is foaf:primaryTopic of	wikipedia-en:Multi-task_learning

Faceted Search & Find service v1.17_git147 as of Sep 06 2024

Alternative Linked Data Documents: ODE Content Formats:

RDF

ODATA

Microdata

About

OpenLink Virtuoso version 08.03.3331 as of Sep 2 2024, on Linux (x86_64-generic-linux-glibc212), Single-Server Edition (378 GB total memory, 53 GB memory in use)
Data on this page belongs to its respective rights holders.
Virtuoso Faceted Browser Copyright © 2009-2024 OpenLink Software