Automated curricula through setter-solver interactions

TL;DR

This paper introduces an automatic curriculum generation method for goal-conditioned reinforcement learning agents in dynamic environments, addressing sparse rewards by considering goal validity, feasibility, and coverage.

Contribution

It presents a novel setter-solver framework for automatic curriculum creation in environments with varying goals, a first in goal-conditioned settings with changing goal sets.

Findings

Effective curriculum generation in 2D and 3D environments.

Guidance towards desired goal distributions improves learning.

Addresses challenges of sparse rewards and goal variability.

Abstract

Reinforcement learning algorithms use correlations between policies and rewards to improve agent performance. But in dynamic or sparsely rewarding environments these correlations are often too small, or rewarding events are too infrequent to make learning feasible. Human education instead relies on curricula--the breakdown of tasks into simpler, static challenges with dense rewards--to build up to complex behaviors. While curricula are also useful for artificial agents, hand-crafting them is time consuming. This has lead researchers to explore automatic curriculum generation. Here we explore automatic curriculum generation in rich, dynamic environments. Using a setter-solver paradigm we show the importance of considering goal validity, goal feasibility, and goal coverage to construct useful curricula. We demonstrate the success of our approach in rich but sparsely rewarding 2D and 3D environments, where an agent is tasked to achieve a single goal selected from a set of possible goals that varies between episodes, and identify challenges for future work. Finally, we demonstrate the value of a novel technique that guides agents towards a desired goal distribution. Altogether, these results represent a substantial step towards applying automatic task curricula to learn complex, otherwise unlearnable goals, and to our knowledge are the first to demonstrate automated curriculum generation for goal-conditioned agents in environments where the possible goals vary between episodes.