1

Currently I am using RL agent DQN to predict action and update the action value function. But if I have a constraint to run a particular action for n times, can I have a constraint in DQN specified for the agent to take action? If yes, How can I possibly do it (using TF)?

Thank you in advance

python_interest
  • 874
  • 1
  • 9
  • 27

1 Answers1

0

Just repeat the action for n steps. There are wrappers like SuperSuit if you're using a gym environment. SuperSuit provides the sticky_actions_v0 or frame_skip_v0 for this.

tnfru
  • 296
  • 1
  • 10