On the Benefit of Optimal Transport for Curriculum Reinforcement Learning
This work addresses the challenge of generating effective curricula for reinforcement learning, which is incremental as it builds on prior interpolation-based approaches.
The paper tackles the problem of automating curriculum generation in reinforcement learning by framing it as a constrained optimal transport problem between task distributions, resulting in improved performance over existing methods across various tasks.
Curriculum reinforcement learning (CRL) allows solving complex tasks by generating a tailored sequence of learning tasks, starting from easy ones and subsequently increasing their difficulty. Although the potential of curricula in RL has been clearly shown in various works, it is less clear how to generate them for a given learning environment, resulting in various methods aiming to automate this task. In this work, we focus on framing curricula as interpolations between task distributions, which has previously been shown to be a viable approach to CRL. Identifying key issues of existing methods, we frame the generation of a curriculum as a constrained optimal transport problem between task distributions. Benchmarks show that this way of curriculum generation can improve upon existing CRL methods, yielding high performance in various tasks with different characteristics.