Crowdsourcing has emerged as a powerful paradigm for efficiently labeling large datasets and performing various learning tasks, by leveraging crowds of human annotators. When additional information is available about the data, constrained or semi-supervised crowdsourcing approaches that enhance the aggregation of labels from human annotators are well motivated. This work deals with constrained crowdsourced classification with instance-level constraints, that capture relationships between pairs of data. A Bayesian algorithm based on variational inference is developed, and its quantifiably improved performance, compared to unsupervised crowdsourcing, is analytically and empirically validated on several crowdsourcing datasets.
|Original language||English (US)|
|Title of host publication||Machine Learning and Knowledge Discovery in Databases. Research Track - European Conference, ECML PKDD 2021, Proceedings|
|Editors||Nuria Oliver, Fernando Pérez-Cruz, Stefan Kramer, Jesse Read, Jose A. Lozano|
|Publisher||Springer Science and Business Media Deutschland GmbH|
|Number of pages||17|
|State||Published - 2021|
|Event||European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2021 - Virtual, Online|
Duration: Sep 13 2021 → Sep 17 2021
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2021|
|Period||9/13/21 → 9/17/21|
Bibliographical noteFunding Information:
Work in this paper was supported by NSF grant 1901134.
© 2021, Springer Nature Switzerland AG.
- Ensemble learning
- Variational inference