Double or nothing for spin filters

Double or Nothing: Multiplicative Incentive Mechanisms for Crowdsourcing

Many fields of science and engineering, ranging from predicting protein structures to building machinetranslation systems, require large amounts of labeled data. These labeling tasks have traditionally beenperformed by experts; the limited pool of experts would limit the size of the datasets, and make the processslow and expensive. In recent years, there is a rapidly increasing ...

Inhaled corticosteroid dosing: double for nothing?

Two recent trials from the National Heart, Lung, and Blood Institute's asthma clinical trials networks raise a concern about using double the dose of an inhaled corticosteroid (ICS) as a positive control arm in clinical trials of add-on therapy. The literature evaluating the response to doubling the dose of an ICS is briefly reviewed. The vast majority of studies do not demonstrate a significan...

Take-all or nothing

Take-all disease of Poaceae is caused by Gaeumannomyces graminis (Magnaporthaceae). Four varieties are recognised in G. graminis based on ascospore size, hyphopodial morphology and host preference. The aim of the present study was to clarify boundaries among species and varieties in Gaeumannomyces by combining morphology and multi-locus phylogenetic analyses based on partial gene sequences of I...

All-or-Nothing Monitoring

A principal can observe both the output and input of an agent who works at a job involving multiple tasks. We provide a simple theory that explains why it may be optimal for the principal to use only output-based incentive contract, even though the principal can monitor the agent’s actions perfectly in all but one task and knows exactly which action is optimal for each task. (JEL: D82, M5, J33)

Overwriting : All or Nothing ?