D

Dqn LunarLander V2

Developed by araffin
This is a DQN agent trained using the stable-baselines3 library to solve reinforcement learning tasks in the LunarLander-v2 environment.
Downloads 54
Release Time : 5/5/2022

Model Overview

This model is based on the Deep Q-Network (DQN) algorithm, specifically designed to solve landing control problems in the LunarLander-v2 environment.

Model Features

Stable Training
Implemented using the stable-baselines3 library, providing a stable training process and reliable performance.
Efficient Exploration
Utilizes optimized exploration strategies, completing exploration within 40,000 timesteps.
Dual-Layer Network Architecture
Employs a 256x256 dual-layer neural network structure, balancing model capacity and training efficiency.

Model Capabilities

Reinforcement Learning
Continuous Control
Environment Interaction
Decision Making

Use Cases

Game AI
Lunar Lander Control
Controls a virtual lunar lander to safely land in a designated area.
Average reward 280.22±13.03
Educational Demonstration
Reinforcement Learning Teaching Example
Serves as a teaching case for deep reinforcement learning.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase