Active Learning from Crowds

By Rómer Rosales, glenn. fung, Jennifer G. Dy. International Conference on Machine Learning, 2011

354 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/a3981ab7727e0606620c1b4fae3def5e3904c010

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Cleaning Crowdsourced Labels Using Oracles For Statistical Classification (2018, background): “There are three research topics in crowdsourcing related to our work: task assignment [4, 6, 13, 17, 26, 31, 46], truth inference [12,16,19,33,45], and active learning (from crowds) [18, 24, 28, 43].”
On-the-Job Learning with Bayesian Decision Theory (2015, unknown): “informative examples to build a classiﬁer. Online active learning [8, 9, 10] performs active learning in the online setting. Several authors have also considered using crowd workers as a noisy oracle [20, 21, 22, 23]. It differs from our setup in that it assumes that labels can only be observed afte”
A brief introduction to weakly supervised learning (2018, background): “Though they are discussed separately, in practice they often occur simultaneously, as illustrated in Figure 1, and there are some relevant studies on such “mixed” cases [60,66,88].”
Challenges in Data Crowdsourcing (2016, background): “There has been initial work in this space [35], [62], but much more remains to be done.”
Annotating Twitter Data from Vulnerable Populations: Evaluating Disagreement Between Domain Experts and Graduate Student Annotators (2018, background): “Essentially, this would mean adopting an active learning paradigm for selectively collecting annotations, similar to [24], but instead of focusing on detectors, expert annotations would be selected in order to train annotators or content moderators.”
Crowdsourced Data Management: A Survey (2016, background): “Many machine learning [54], [55], [64] and active learning techniques [37], [82], [130], [140] aim to assign a set of tasks to workers that are”
Artificial Intelligence and Collective Intelligence (2014, background): “(Yan et al. 2011) assume worker skills are known and adaptively select the most confident worker for a given question (using a model based on question features).”
Pairwise ranking aggregation in a crowdsourced setting (2013, background): “Yan et al. [30] proposed an active learning strategy for binary classification with a crowd.”