Timeline for The effect of policy parameter on the action and the state distribution in policy gradient method for episodic tasks
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 18, 2018 at 7:24 | vote | accept | Ali Ghghgh | ||
| Dec 18, 2018 at 19:38 | |||||
| Dec 18, 2018 at 7:22 | comment | added | shimao | @AliGh if you mean "the state" by environment, then yes. If you mean the underlying MDP, then no, an agent cannot change that. | |
| Dec 18, 2018 at 7:19 | comment | added | shimao | Consider an environment with three states, left, middle, and right, and each episode starts with the agent in the middle state. The agent has two actions, L and R, which move the agent to the left / right state respectively. The state distribution for an policy which always takes the L action is obviously different from the state distribution of a policy which always takes the R action. | |
| Dec 18, 2018 at 7:13 | comment | added | Ali Ghghgh | Based on what you say, the agent should have the power to change the environment. Can you give me an example or reference for more details? | |
| Dec 18, 2018 at 6:58 | comment | added | shimao | "Because state distribution is about environment". No, the state distribution is a function of both the environment and the agent | |
| Dec 18, 2018 at 6:55 | comment | added | Ali Ghghgh | But my problem is that what is the point of the author when it says the effect of policy on the state distribution is unknown? Because state distribution is about environment and policy is what agent does, so affecting the environment by the agent cannot happen. | |
| Dec 17, 2018 at 5:56 | vote | accept | Ali Ghghgh | ||
| Dec 17, 2018 at 5:57 | |||||
| Dec 17, 2018 at 4:18 | history | answered | shimao | CC BY-SA 4.0 |