DQN-SpaceInvadersNoFrameskip-v4 Open-source Game Agent - Freely Deploy and Enjoy a Specific Game Environment

Dqn SpaceInvadersNoFrameskip V4

Developed by epsil

This is a reinforcement learning agent based on the DQN algorithm, specifically designed for gameplay in the SpaceInvadersNoFrameskip-v4 environment.

Video Processing #Atari game reinforcement learning #Deep Q-Network #High frame rate environment

Downloads 15

Release Time : 6/8/2022

Model Overview

The model was trained using the stable-baselines3 library and RL Zoo framework, capable of playing Space Invaders with an average reward of 637.50 +/- 139.13.

Model Features

Stable Training Framework

Trained using the stable-baselines3 library and RL Zoo framework to ensure training stability

Efficient Exploration Strategy

Employs a gradually decaying exploration rate strategy, transitioning from initial exploration to a stable final policy

Frame Stacking Processing

Uses 4-frame stacking technique to process game screens, helping the agent understand dynamic environments

Model Capabilities

Atari game playing

Reinforcement learning decision-making

Game screen understanding

Use Cases

Game AI

Space Invaders Game AI

This model can autonomously play Space Invaders

Average reward reaches 637.50 +/- 139.13

Reinforcement Learning Research

DQN Algorithm Benchmarking

Can serve as a performance benchmark for DQN algorithm on Atari games

🚀 DQN Agent for SpaceInvadersNoFrameskip-v4

This is a trained DQN agent for the SpaceInvadersNoFrameskip-v4 game, leveraging the stable-baselines3 library and the RL Zoo to achieve effective reinforcement learning.

🚀 Quick Start

This is a trained model of a DQN agent playing SpaceInvadersNoFrameskip-v4 using the stable-baselines3 library and the RL Zoo. The RL Zoo is a training framework for Stable Baselines3 reinforcement learning agents, with hyperparameter optimization and pre-trained agents included.

📦 Installation

RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo
SB3: https://github.com/DLR-RM/stable-baselines3
SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib

💻 Usage Examples

Basic Usage

# Download model and save it into the logs/ folder
python -m utils.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga epsil -f logs/
python enjoy.py --algo dqn --env SpaceInvadersNoFrameskip-v4  -f logs/

Advanced Usage

python train.py --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
# Upload the model and generate video (when possible)
python -m utils.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga epsil

🔧 Technical Details

Hyperparameters

OrderedDict([('batch_size', 32),
             ('buffer_size', 100000),
             ('env_wrapper',
              ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
             ('exploration_final_eps', 0.01),
             ('exploration_fraction', 0.1),
             ('frame_stack', 4),
             ('gradient_steps', 1),
             ('learning_rate', 0.0001),
             ('learning_starts', 100000),
             ('n_timesteps', 10000000.0),
             ('optimize_memory_usage', True),
             ('policy', 'CnnPolicy'),
             ('target_update_interval', 1000),
             ('train_freq', 4),
             ('normalize', False)])

Performance Metrics

Property	Details
Model Type	DQN
Training Data	SpaceInvadersNoFrameskip-v4
Mean Reward	637.50 +/- 139.13

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご