Let the Model Decide its Curriculum for Multitask Learning

Anonymous

Abstract: Curriculum learning strategies in prior multi-task learning approaches arrange datasets in a difficulty hierarchy either based on human perception or by exhaustively searching the optimal arrangement. However, human perception of difficulty may not always correlate well with machine interpretation leading to poor performance and exhaustive search is computationally expensive. Addressing these concerns, we propose two classes of techniques to arrange training instances into a learning curriculum based on difficulty scores computed via model-based approaches. The two classes i.e Dataset-level and Instance-level differ in the granularity of arrangement. We conduct comprehensive experiments with $12$ datasets and show that instance-level and dataset-level techniques lead to an average performance improvement of $4.17\%$ and $3.15\%$ over their respective baseline methods. Furthermore, we find that most of this improvement comes from correctly answering the difficult instances, implying a greater efficacy of our techniques on difficult tasks.

Paper Type: short