Robust crowdsourcing-based linear regression

S. Abbaasi, Majeed Mohammadi, E.S. Davodly

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

In most machine learning problems, the labeling of the training data is an expensive or even impossible task. Crowdsourcing-based learning uses uncertain labels from many non-expert annotators instead of one reference label. Crowdsourcing based linear regression is an efficient way for function estimation when many labels are available for each instance. However, methods in literature have a poor performance against large noise and outliers in labels. To tackle this problem, we proposed a novel robust crowdsourcing-based linear regression derived from information theoretic learning. The proposed problem is not convex, but it can be efficiently solved by half quadratic programming. The proposed model has a close relation with weighted crowdsourcing-based linear regression, in which labels of each annotator weight adaptively and iteratively. The Performance of the proposed method evaluated on several artificial data sets in different circumstances. Experimental Results demonstrate the efficacy and robustness of the proposed method.

Original languageEnglish
Title of host publicationProceedings of 6th International Conference on Computer and Knowledge Engineering, ICCKE 2016
PublisherIEEE
Pages141-146
ISBN (Electronic)9781509035861
DOIs
Publication statusPublished - 2016
Event6th International Conference on Computer and Knowledge Engineering, ICCKE 2016 - Mashhad, Iran, Islamic Republic of
Duration: 20 Oct 2016 → …

Conference

Conference6th International Conference on Computer and Knowledge Engineering, ICCKE 2016
CountryIran, Islamic Republic of
CityMashhad
Period20/10/16 → …

Keywords

  • crowdsourcing
  • information theoretic learning
  • outlier labels
  • robust linear regression

Fingerprint Dive into the research topics of 'Robust crowdsourcing-based linear regression'. Together they form a unique fingerprint.

Cite this