Do you want to estimate causal effects for a small set of target variables without knowing the causal graph, but discovering it takes too long? Now you can get adjustment sets in a SNAP🫰accepted at #aistats2025! 📜 arxiv.org/abs/2502.07857 🧩 matyasch.github.io/snap/ 🧵 1/10 - ThreadSky

matyasch.bsky.social • 19 days ago

Do you want to estimate causal effects for a small set of target variables without knowing the causal graph, but discovering it takes too long? Now you can get adjustment sets in a SNAP🫰accepted at #aistats2025!

📜 https://arxiv.org/abs/2502.07857
🧩 https://matyasch.github.io/snap/
🧵 1/10

Comments

matyasch.bsky.social•19 days ago

2/10 Discovering causal relations can help us estimate causal effects, but it is expensive 📈 If we are only interested in estimating the causal effects between a few target variables, can we instead only discover a subgraph that includes these targets and their adjustment sets?

matyasch.bsky.social•19 days ago

3/10 We formalize this as the task of “targeted causal effect estimation with an unknown graph”, which focuses on identifying causal effects between a small set of target variables in a ✨computationally and statistically efficient way✨

matyasch.bsky.social•19 days ago

4/10 To solve this task, we show that only possible ancestors of the targets are required to identify their causal relationships and efficient adjustment sets💡 Driven by this, we propose SNAP to progressively prune non-ancestors, leading to much fewer higher order CI tests.

matyasch.bsky.social•19 days ago

5/10 SNAP is straightforward to combine with readily available causal discovery algorithms 🧩 We can simply stop it at any maximum iteration k and run another algorithm on the remaining variables. We refer to this approach as prefiltering with SNAP(k).

matyasch.bsky.social•19 days ago

6/10 We can also run SNAP until completion, to obtain a stand-alone causal discovery algorithm, called SNAP(∞). SNAP(∞) is sound and complete over the possible ancestors of targets ✅ Thus, unlike previous work on local causal discovery, it finds efficient adjustment sets.

matyasch.bsky.social•19 days ago

7/10 SNAP(∞) consistently ranks among the best in the number of CI tests and computation time across all domains, while maintaining a comparable intervention distance. In contrast, other methods vary in performance depending on the setting 🚀

Comments

Posting Rules

Reply