Active Multitask Learning with Committees

TL;DR

This paper introduces an active multitask learning algorithm that uses committees to share information across related tasks, reducing labeling costs while maintaining high accuracy.

Contribution

It proposes a novel committee-based approach for active multitask learning that enhances knowledge transfer and reduces annotation efforts.

Findings

Significant reduction in query requests during training

Maintains high test accuracy with fewer labels

Improves performance on benchmark datasets

Abstract

The cost of annotating training data has traditionally been a bottleneck for supervised learning approaches. The problem is further exacerbated when supervised learning is applied to a number of correlated tasks simultaneously since the amount of labels required scales with the number of tasks. To mitigate this concern, we propose an active multitask learning algorithm that achieves knowledge transfer between tasks. The approach forms a so-called committee for each task that jointly makes decisions and directly shares data across similar tasks. Our approach reduces the number of queries needed during training while maintaining high accuracy on test data. Empirical results on benchmark datasets show significant improvements on both accuracy and number of query requests.