Mastering Rate based Curriculum Learning
This work addresses a specific bottleneck in automatic curriculum learning for machine learning practitioners, offering an incremental improvement over prior approaches.
The paper tackles the low sample efficiency of learning progress-based curriculum learning algorithms by proposing a new algorithm based on mastering rate, which significantly outperforms existing methods.
Recent automatic curriculum learning algorithms, and in particular Teacher-Student algorithms, rely on the notion of learning progress, making the assumption that the good next tasks are the ones on which the learner is making the fastest progress or digress. In this work, we first propose a simpler and improved version of these algorithms. We then argue that the notion of learning progress itself has several shortcomings that lead to a low sample efficiency for the learner. We finally propose a new algorithm, based on the notion of mastering rate, that significantly outperforms learning progress-based algorithms.