Skip to content

Repository files navigation

SnakeAI3.mp4
car-race-ai-vlc.mp4
lunar_lander1.mp4
simplescreenrecorder-2026-05-07_13.25.13.mp4
ContinuousMountainCar_BestEra.mp4
simplescreenrecorder-2026-06-28_18.47.41.mp4

Ghost ReinforcementLearning #GhostRL

Agentic AI build, train and evaluation utilizing Windows and Ubuntu Linux, TensorFlow and Proximal Policy Optimization (PPO) #OpenSource #TheEnd

Snake_AI agent performance after 10000000 training timesteps: https://github.com/user-attachments/assets/a74d5d9e-dbca-4391-b6fe-d3474a153358

Car_Race_AI agent performance on multiple tracks and different driving techniques after 16558080 training timesteps: https://github.com/user-attachments/assets/7edd7a74-66a0-4b43-9e14-8072bf7aeb84

Discrete_Lunar_Lander_AI performance after 15000000 training timesteps: https://github.com/user-attachments/assets/5222bfa1-734f-4da6-9f07-14dd9b0b5263

Taxi_AI agent after 40000000 training timesteps: https://github.com/user-attachments/assets/64b71fc5-3b62-48ec-acc6-d07209c339dc

ContinuousMountainCar_BestEra_AI performance after 18000000 training timesteps: https://github.com/user-attachments/assets/87f4405c-7079-4d16-8339-32585a644967

DiscreteMountainCar_NoReward_AI performance after 32000000 training timesteps: https://github.com/user-attachments/assets/684ff17f-9ba0-423a-80c6-96fe4d073e22

About

Agentic AI build, train and action

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Used by

Contributors

Languages