Explanation:

1. Training an agent in atari breakout environment with deep Q-learning (DQN) to play Atari game.

2. Training an agent in pong environment with deep Q-learning (DQN) to play ping pong game.

3. Training an agent in continuous pendulum environment using Deep Deterministic Policy Gradient (DDPG). The goal of the agent is to Keep the pendulum upright.

4. Training an agent in Cart Pole environment using Proximal Policy Optimization (PPO) as well as Actor-Critic algorithm. The goal of the agent is to balance the pole upright and keep the cart within the track boundaries.