Approval Voting and Incentives in Crowdsourcing

Authors: Nihar Shah, Dengyong Zhou, Yuval Peres

ICML 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We also conduct preliminary empirical studies on Amazon Mechanical Turk which validate our approach.
Researcher Affiliation Collaboration Nihar B. Shah NIHAR@EECS.BERKELEY.EDU University of California, Berkeley, CA 94720 Dengyong Zhou DENZHO@MICROSOFT.COM Microsoft Research, Redmond, WA 98052 Yuval Peres PERES@MICROSOFT.COM Microsoft Research, Redmond, WA 98052
Pseudocode Yes Algorithm 1 Incentive mechanism for approval voting
Open Source Code No The paper states 'The entire data related to the experiments is available on the website of the first author' but does not mention code for the methodology.
Open Datasets No The paper states 'The entire data related to the experiments is available on the website of the first author' but does not provide a specific link, DOI, repository, or formal citation for a publicly available dataset. It refers to data collected by the authors.
Dataset Splits No The paper does not explicitly provide details about training, validation, or test dataset splits.
Hardware Specification No The paper mentions Amazon Mechanical Turk as the platform for crowdsourcing but does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments or models.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes In each experiment, every worker was assigned one of four mechanisms uniformly at random. The mechanisms were executed as a bonus payment based on the evaluation of the worker s performance on the gold standard questions, on top of a guaranteed payment of 10 cents. The four mechanisms tested were: Single-selection interface with additive payments: The bonus starts at zero and is increased by a fixed amount for every correct answer. Skip-based single-selection interface with multiplicative payments (Shah & Zhou, 2014): The bonus starts at a certain positive value, is reduced by a certain fraction for each skipped question, and becomes zero in case of an incorrect answer. Approval-voting interface with a fixed payment: The bonus is fixed. Approval-voting interface with the payment defined in Algorithm 1.