lexiconium / cartpole

Cartpole

Simple Open-AI Gym CartPole experiments.

Available Algorithms

Deeq Q-Network

DQN: Playing Atari with Deep Reinforcement Learning

python dqn.py

Advantage Actor-Critic

A2C: Asynchronous Methods for Deep Reinforcement Learning

python a2c.py

Proximal Policy Optimization

PPO: Proximal Policy Optimization Algorithms

python ppo.py

Additional Arguments

--env ENV
--batch_size BATCH_SIZE
--num_episodes NUM_EPISODES
--update_interval UPDATE_INTERVAL
--learning_rate LEARNING_RATE
--weight_decay WEIGHT_DECAY
--gamma GAMMA
--epsilon EPSILON
--seed SEED