Deep Reinforcement Learning Projects
POCA Agent playing SoccerTwos
This project uses a POCA agent to simulate strategic play in SoccerTwos, showcasing multi-agent coordination and competitive soccer simulation using reinforcement learning techniques.
View on Hugging FaceA2C Agent playing PandaReachDense-v2
An A2C agent learns to control a Panda robotic arm in the PandaReachDense-v2 environment, focusing on precision and dense reward signals.
View on Hugging FaceA2C Agent playing AntBulletEnv-v0
In this project, an A2C agent is trained in the AntBulletEnv-v0 environment, acquiring complex locomotion and coordination skills.
View on Hugging FaceReinforce Agent playing CartPole-v1
A REINFORCE agent tackles the classic CartPole-v1 challenge, learning to balance the pole using policy gradient methods.
View on Hugging FaceDQN Agent playing SpaceInvadersNoFrameskip-v4
This project demonstrates a Deep Q-Network (DQN) agent mastering the classic SpaceInvaders game using deep learning strategies.
View on Hugging FaceQ-Learning Agent playing FrozenLake-v1
A Q-learning agent is implemented in the FrozenLake-v1 environment, demonstrating value-based learning in a stochastic setting.
View on Hugging FacePPO Agent playing LunarLander-v2
A Proximal Policy Optimization (PPO) agent is trained for the LunarLander-v2 challenge, balancing exploration and fine-tuning for optimal performance.
View on Hugging Face