In this project, I use reinforcement learning (RL) to train two agents playing tennis in an environment similar to Unity's Tennis environment.
In this environment, two agents control rackets to bounce a ball over a net. If an agent hits the ball over the net, it receives a reward of +0.1. If an agent lets a ball hit the ground or hits the ball out of bounds, it receives a reward of -0.01. Thus, the goal of each agent is to keep the ball in play.
The observation space consists of 8 variables corresponding to the position and velocity of the ball and racket. Each agent receives its own, local observation. Two continuous actions are available, corresponding to movement toward (or away from) the net, and jumping.
The task is episodic, and in order to solve the environment, the agents must get an average score of +0.5 (over 100 consecutive episodes, after taking the maximum over both agents). Specifically,
- After each episode, we add up the rewards that each agent received (without discounting), to get a score for each agent.
- This yields 2 (potentially different) scores. We then take the maximum of these 2 scores.
- This yields a single score for each episode.
The environment is considered solved, when the average (over 100 episodes) of those scores is at least +0.5.Consequently, the environment can be categoriezed as an episodic, multi-agent, continous control problem.
cd /python
pip install .
- Run
python3 watch_trained_agent.pyto see the trained agent in action. - Run
python3 watch_random_agent.pyto see an untrained agent performing random actions. - Run
python3 train_control_agent.pyto re-train the agent. model.pydefines the actor and critic network architectures.agent.pydefines the MADDPG agent class.*.pthfiles each contain the saved network weights after training.
Details on the training algorithm, network architecture and hyperparameters are discussed in the
The implemented RL algorithm is able to solve the environment in about 1800 episodes:

- Perform a systematic hyperparameter optimization study to improve convergence characteristics.
- Benchmark against other algorithms, such as TRPO (Trust Region Policy Optimization), PPO (Proximal Policy Optimization) or D4PG (Distributed Distributional Deterministic Policy Gradients), which may obtain more robust results.
