1. सीखें
    2. /
  2. पाठ्यक्रम
    3. /
  3. Deep Reinforcement Learning in Python
Connected

अभ्यास

Actor Critic loss calculations

As a final step before you can train your agent with A2C, write a calculate_losses() function which returns the losses for both networks.

For reference, these are the expressions for the actor and critic loss functions respectively:

निर्देश

100 XP
  • Calculate the TD target.
  • Calculate the loss for the Actor network.
  • Calculate the loss for the Critic network.