6
$\begingroup$

I'm training both DQN and double DQN in the same environment, but DQN performs significantly better than double DQN. As I've seen in the double DQN paper, double DQN should perform better than DQN. Am I doing something wrong or is it possible?

$\endgroup$
2
  • 1
    $\begingroup$ Hello. I am facing the same issue. What explanation finally did you give ? $\endgroup$ – ddaedalus Jan 10 at 12:18
  • $\begingroup$ I have the same result, DQN performs better than Double DQN. I m try to understand why. $\endgroup$ – Mustafa Jan 25 at 16:28
1
$\begingroup$

There is no thorough proof, theoretical or experimental that Double DQN is better then vanilla DQN. There are a lot of different tasks, paper and later experiments only explore some of them. What practitioner can take out of it is that on some tasks DDQN is better. That's the essence of Deep Mind's "Rainbow" approach - drop a lot of different methods into bucket and take best results.

$\endgroup$
2
  • 1
    $\begingroup$ Can you give some references where DQN outperforms DoubleDQN? $\endgroup$ – ddaedalus Jan 10 at 12:13
  • 1
    $\begingroup$ I have a problem where my DQN is better than DoubleDQN and I cannot give a valid explanation on why. $\endgroup$ – ddaedalus Jan 10 at 12:14
0
$\begingroup$

That may happen when the value of the state is bad. You can find the example and explain about that in the link below.

See this:https://medium.freecodecamp.org/improvements-in-deep-q-learning-dueling-double-dqn-prioritized-experience-replay-and-fixed-58b130cc5682

$\endgroup$
6
  • 1
    $\begingroup$ What do you mean by "the value of the state is bad"? As I'm using an OpenAI-Gym environment, the value of the state is just the observation that I'm getting from it. $\endgroup$ – Angelo Apr 9 '19 at 7:53
  • $\begingroup$ @Angelo you can read the blog above to understand the answer . As you calculate all actions at one state but all these actions do not affect the environment in a relevant way. $\endgroup$ – i_th Apr 9 '19 at 8:11
  • $\begingroup$ @Angelo In most states, the choice of action has no effect. $\endgroup$ – i_th Apr 9 '19 at 8:37
  • 2
    $\begingroup$ We prefer answers that contain the necessary info to answer the question, rather than answers that just link to another external resource, so I suggest that you provide the necessary info from that article in your answer. $\endgroup$ – nbro Nov 4 '20 at 21:23
  • $\begingroup$ In your link, it says that in DuelingDQN session, not in DoubleDQN as asked. $\endgroup$ – ddaedalus Jan 10 at 12:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.