ELSIM: End-to-end learning of reusable skills through intrinsic motivation

TL;DR

This paper introduces ELSIM, a hierarchical reinforcement learning framework that autonomously learns transferable skills through intrinsic motivation, enhancing exploration and performance in complex environments.

Contribution

ELSIM presents a novel end-to-end architecture combining mutual information and curriculum learning to develop an explorable skill tree for reinforcement learning agents.

Findings

Skills are transferable across tasks.

Improves exploration in sparse reward settings.

Scales to complex MuJoCo environments.

Abstract

Taking inspiration from developmental learning, we present a novel reinforcement learning architecture which hierarchically learns and represents self-generated skills in an end-to-end way. With this architecture, an agent focuses only on task-rewarded skills while keeping the learning process of skills bottom-up. This bottom-up approach allows to learn skills that 1- are transferable across tasks, 2- improves exploration when rewards are sparse. To do so, we combine a previously defined mutual information objective with a novel curriculum learning algorithm, creating an unlimited and explorable tree of skills. We test our agent on simple gridworld environments to understand and visualize how the agent distinguishes between its skills. Then we show that our approach can scale on more difficult MuJoCo environments in which our agent is able to build a representation of skills which improve over a baseline both transfer learning and exploration when rewards are sparse.