Training Neural Nets to Aggregate Crowdsourced Responses
Abstract
We propose a new method for aggregating crowdsourced responses, based on a deep neural network. Once trained, the aggregator network gets as inputs the responses of multiple participants to a the same set of questions, and outputs its prediction for the correct response to each question. We empirically evaluate our approach on a dataset of responses to a standard IQ questionnaire, and show it outperforms existing state-of-the-art methods.
Cite
Text
Gaunt et al. "Training Neural Nets to Aggregate Crowdsourced Responses." Conference on Uncertainty in Artificial Intelligence, 2016.Markdown
[Gaunt et al. "Training Neural Nets to Aggregate Crowdsourced Responses." Conference on Uncertainty in Artificial Intelligence, 2016.](https://mlanthology.org/uai/2016/gaunt2016uai-training/)BibTeX
@inproceedings{gaunt2016uai-training,
title = {{Training Neural Nets to Aggregate Crowdsourced Responses}},
author = {Gaunt, Alex and Borsa, Diana and Bachrach, Yoram},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {2016},
url = {https://mlanthology.org/uai/2016/gaunt2016uai-training/}
}