1. Learn
    2. /
  2. Courses
    3. /
  3. Deep Reinforcement Learning in Python
Connected

Exercise

Training the barebone DQN

It's time to train a Barebone DQN algorithm in the Lunar Lander environment. Keep in mind this is still a bare algorithm, so the performance won't be great, but you'll improve on it later.

Think of it as the first step towards getting your Lunar Lander to land on the Moon!

The q_network instance that you defined earlier is available to you.

Throughout the exercises in this course, your Python environment also comes with a describe_episode() function to print some information at the end of each episode about how the agent has fared.

Instructions

100 XP
  • Select the agent's action in the inner loop.
  • Calculate the loss.
  • Perform a gradient descent step to update the network weights.