Automated Curriculum Learning by Rewarding Temporally Rare Events

TL;DR

This paper introduces a novel reward shaping method called Rarity of Events (RoE) that encourages exploration by rewarding less frequent events, enabling RL agents to learn effectively in complex, sparse-reward environments without manual reward design.

Contribution

The paper presents a simple, general approach for automated curriculum learning by rewarding rare events, removing the need for manual reward shaping in complex RL tasks.

Findings

RoE enables agents to succeed in VizDoom without extrinsic rewards.

RoE results in more versatile and adaptable policies.

The approach is effective in environments with sparse rewards and many event types.

Abstract

Reward shaping allows reinforcement learning (RL) agents to accelerate learning by receiving additional reward signals. However, these signals can be difficult to design manually, especially for complex RL tasks. We propose a simple and general approach that determines the reward of pre-defined events by their rarity alone. Here events become less rewarding as they are experienced more often, which encourages the agent to continually explore new types of events as it learns. The adaptiveness of this reward function results in a form of automated curriculum learning that does not have to be specified by the experimenter. We demonstrate that this \emph{Rarity of Events} (RoE) approach enables the agent to succeed in challenging VizDoom scenarios without access to the extrinsic reward from the environment. Furthermore, the results demonstrate that RoE learns a more versatile policy that adapts well to critical changes in the environment. Rewarding events based on their rarity could help in many unsolved RL environments that are characterized by sparse extrinsic rewards but a plethora of known event types.