Timezone: »
Most existing works on crowdsourcing assume that the workers follow the Dawid-Skene model, or the one-coin model as its special case, where every worker makes mistakes independently of other workers and with the same error probability for every task. We study a significant extension of this restricted model. We allow almost half of the workers to deviate from the one-coin model and for those workers, their probabilities of making an error to be task-dependent and to be arbitrarily correlated. In other words, we allow for arbitrary adversaries, for which not only error probabilities can be high, but which can also perfectly collude. In this adversarial scenario, we design an efficient algorithm to consistently estimate the workers’ error probabilities.
Author Information
Matthäus Kleindessner (Rutgers University)
Pranjal Awasthi (Rutgers University)
Related Events (a corresponding poster, oral, or spotlight)
-
2018 Oral: Crowdsourcing with Arbitrary Adversaries »
Wed. Jul 11th 09:00 -- 09:20 AM Room A4
More from the Same Authors
-
2020 Poster: Adversarial Learning Guarantees for Linear Hypotheses and Neural Networks »
Pranjal Awasthi · Natalie Frank · Mehryar Mohri -
2019 Poster: Fair k-Center Clustering for Data Summarization »
Matthäus Kleindessner · Pranjal Awasthi · Jamie Morgenstern -
2019 Poster: Guarantees for Spectral Clustering with Fairness Constraints »
Matthäus Kleindessner · Samira Samadi · Pranjal Awasthi · Jamie Morgenstern -
2019 Oral: Guarantees for Spectral Clustering with Fairness Constraints »
Matthäus Kleindessner · Samira Samadi · Pranjal Awasthi · Jamie Morgenstern -
2019 Oral: Fair k-Center Clustering for Data Summarization »
Matthäus Kleindessner · Pranjal Awasthi · Jamie Morgenstern -
2018 Poster: Clustering Semi-Random Mixtures of Gaussians »
Aravindan Vijayaraghavan · Pranjal Awasthi -
2018 Oral: Clustering Semi-Random Mixtures of Gaussians »
Aravindan Vijayaraghavan · Pranjal Awasthi