|
Activity Number:
|
41
- Topics on Bayesian Inference
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
IMS
|
|
Abstract #312506
|
|
|
Title:
|
Crowdsourcing: Beyond Dawid-Skene Model
|
|
Author(s):
|
Tony Cai and Ran Chen*
|
|
Companies:
|
University of Pennsylvania and Department of Statistics, Wharton, Upenn
|
|
Keywords:
|
Crowdsourcing;
Ensemble Learning;
Machine Learning;
Unsupervised Learning
|
|
Abstract:
|
Crowdsourcing has a wide range of applications. Many competing models and methods have been proposed. Those models, however, are too restrictive in different ways. In this work, we introduce an approximately low rank model that encompasses most of the existing models, capturing the important aspects of the real world problem of crowdsourcing. We propose a novel estimator and show its performance under difficulty weighted hamming loss. We also establish a lower bound to set a baseline.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2020 program
|
