Episode Q0 increases exponentially
18 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Can anyone explain why episode Q0 in RL increases exponentially after convergence of reward to a suboptimal policy?
0 comentarios
Respuestas (1)
Emmanouil Tzorakoleftherakis
el 16 de Feb. de 2021
Hello,
Please take a look at this answer for some suggestions. Normalizing observations, rewards, and actions can also help avoid situations like these.
Hope this helps
1 comentario
Ver también
Categorías
Más información sobre Training and Simulation en Help Center y File Exchange.
Productos
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!