Investigating Reinforcement Learning Agents for Continuous State Space Environments

TL;DR

This paper explores the application of Double Deep Q-learning agents to continuous state space environments, specifically using the LunarLander-v2 environment to find optimal policies.

Contribution

It demonstrates the effectiveness of Double Deep Q-learning in continuous state spaces within the LunarLander-v2 environment, a less explored area.

Findings

Double Deep Q-learning successfully learns policies in LunarLander-v2.

The approach outperforms traditional Q-learning methods.

The study provides insights into reinforcement learning in continuous spaces.

Abstract

Given an environment with continuous state spaces and discrete actions, we investigate using a Double Deep Q-learning Reinforcement Agent to find optimal policies using the LunarLander-v2 OpenAI gym environment.