Timeline for How do we define greedy action for policy improvement for a given stochastic policy?
Current License: CC BY-SA 4.0
2 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 14, 2023 at 15:45 | comment | added | Neil Slater | See my comment on the question. This answer assumes the OP does not want to calculate the greedy action, and is searching for something else. However, the question is asking for the definition of the greedy action given the action value function, which is straightforward. Also, if you are using $\epsilon$-greedy behaviour policy, then the greedy action choice is also used in the policy improvement step. The OP has not returned to the question to clarify, although they are still active on the site. | |
| Nov 14, 2023 at 8:05 | history | answered | cinch | CC BY-SA 4.0 |