Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
| uiai [2026/03/17 01:40] – [Definition: Counterfactual action] pedroortega | uiai [2026/03/17 10:30] (current) – [Definition: Third-party action] pedroortega | ||
|---|---|---|---|
| Line 416: | Line 416: | ||
| Note that $k'$ is determined inside the branch and therefore the length of $\dot{a}_{t+1}$ need not match the length of the on-path $\mathcal{A}$-token written by the agent starting at $k$. | Note that $k'$ is determined inside the branch and therefore the length of $\dot{a}_{t+1}$ need not match the length of the on-path $\mathcal{A}$-token written by the agent starting at $k$. | ||
| + | |||
| + | {{ :: | ||
| **Diagram note.** | **Diagram note.** | ||
| - | The intended picture is that after the shared on-path prefix ending at $a_3,o_3$, the on-path transcript has factual action $a_4$, while the counterfactual | + | The diagram shows that after the shared on-path prefix ending at $a_3,o_3$, the on-path transcript has factual action $a_4$, while the counterfactual |
| ==== Definition: Third-party action ==== | ==== Definition: Third-party action ==== | ||
| Line 448: | Line 450: | ||
| o_t = w\, | o_t = w\, | ||
| $$ | $$ | ||
| + | |||
| + | {{ : | ||
| **Diagram note.** | **Diagram note.** | ||
| - | The intended picture is that inside | + | The diagram illustrates a third party action. Inside |
| It is not hard to see that, for a given potential index $k$, the counterfactual action $\dot{a}_{t+1}$ and the third-party action $\dot{a}_{t+1}$ are the same random block: the difference is only whether the gate sampled $\gamma_k = 1$ (counterfactual, | It is not hard to see that, for a given potential index $k$, the counterfactual action $\dot{a}_{t+1}$ and the third-party action $\dot{a}_{t+1}$ are the same random block: the difference is only whether the gate sampled $\gamma_k = 1$ (counterfactual, | ||