Expected reward blows up while training (DDPG agent, reinforcement learning)
1 Ansicht (letzte 30 Tage)
Ältere Kommentare anzeigen
I am training a DDPG network and after training for around 5000 iterations, the model seems doesnot seem to converge while the expected reward keeps on increasing exponentially. What can be a possible reason and how to solve the issue.
0 Kommentare
Antworten (1)
Emmanouil Tzorakoleftherakis
am 12 Okt. 2020
Bearbeitet: Emmanouil Tzorakoleftherakis
am 12 Okt. 2020
Hello,
This answer may be helpful.
I would make sure your reward signal outputs values that make sense, and also possibly simplify the critic network.
2 Kommentare
Siehe auch
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!