9:00 - that is not how it says in Double DQN paper it says to use target network as (phi 2) so we train only one network(phi 1) and there are total 2 networks not 4 ???
Thank you so much for that great explanation! I definitely smashed that subscribe button B) - im wondering which one to use for predictions, maybe an average of both models? - or pick random from either one, to have some generalization to the predictions?
![HIWWHEE | OHM [Official MV]](http://i.ytimg.com/vi/OJJLtTLqLag/mqdefault.jpg)
9:00 - that is not how it says in Double DQN paper it says to use target network as (phi 2) so we train only one network(phi 1) and there are total 2 networks not 4 ???
This is a brilliant explanation, thank you!
Absolutely fantastic explanation. Thanks a lot.
Very clear and intelligible explanation, thank you!
Good one! Clear and concise explanation. Thanks.
Thank you so much for your helpful teaching!
Excellent explanation.
Thank you so much for that great explanation! I definitely smashed that subscribe button B) - im wondering which one to use for predictions, maybe an average of both models? - or pick random from either one, to have some generalization to the predictions?
Superb!
Great video thank you.
Great explanation!
Well explained, thank you !!
Super explanation
Thank you so much for the explanation!
how are the true values computed in the study?
4:45开始