Active Learning Based on Crowdsourced Data

Abstract

The paper proposes a crowdsourcing-based approach for annotated data acquisition and means to support Active Learning training approach. In the proposed solution, aimed at data engineers, the knowledge of the crowd serves as an oracle that is able to judge whether the given sample is informative or not. The proposed solution reduces the amount of work needed to annotate large sets of data. Furthermore, it allows a perpetual increase in the trained network quality by the inclusion of new samples, gathered after network deployment. The paper also discusses means of limiting network training times, especially in the post-deployment stage, where the size of the training set can increase dramatically. This is done by the introduction of the fourth set composed of samples gather during network actual usage.