AIMar 19, 2018

Automated Curriculum Learning by Rewarding Temporally Rare Events

arXiv:1803.07131v220 citations
Originality Highly original
AI Analysis

This addresses the problem of sparse rewards in complex RL environments for researchers and practitioners, offering an automated curriculum learning approach.

The paper tackles the challenge of designing reward signals in reinforcement learning by proposing a method that rewards events based on their rarity, enabling agents to succeed in challenging VizDoom scenarios without extrinsic rewards and learn versatile policies.

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.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes