AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

Zero-shot causal learning

Hamed Nilforoshan, Michael Moor, Yusuf H Roohani, Yining Chen, Anja Šurina, Michihiro Yasunaga, Sara Oblak, Jure Leskovec

NeurIPS 2023 · Probabilistic and scientific modeling › Causal discovery · approach: Causal effect identification · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
remove meta-learning (train with ERM instead of Reptile) (ablation)CaML · Claims, pancytopenia, unseen drugsRATE @ 0.999: 0.39 -> 0.48Improved
remove meta-learning (train with ERM instead of Reptile) (ablation)CaML · Claims, pancytopenia, unseen drugsRATE @ 0.99: 0.11 -> 0.13No clear effect
replace RA-learner with an alternative pseudo-outcome estimator (ablation)CaML · Claims, pancytopenia, unseen drugsRATE @ 0.999: 0.45 -> 0.48Improved
replace RA-learner with an alternative pseudo-outcome estimator (ablation)CaML · Claims, pancytopenia, unseen drugsRATE @ 0.99: 0.14 -> 0.13No clear effect
remove meta-learning (train with ERM instead of Reptile) (ablation)CaML · LINCSPEHE 50 DEGs: 3.57 -> 3.56No clear effect
remove meta-learning (train with ERM instead of Reptile) (ablation)CaML · LINCSPEHE 20 DEGs: 3.79 -> 3.78No clear effect
replace RA-learner with an alternative pseudo-outcome estimator (ablation)CaML · LINCSPEHE 50 DEGs: 4.28 -> 3.56Improved
replace RA-learner with an alternative pseudo-outcome estimator (ablation)CaML · LINCSPEHE 20 DEGs: 4.60 -> 3.78Improved
use pseudo-outcomes (CaML) instead of meta-learning Y(w) and Y(0) directly (T-learner w/ meta-learning)meta-learned CATE model · Claims, pancytopenia, unseen drugsRATE @ 0.999: 0.40 -> 0.48Improved
use pseudo-outcomes (CaML) instead of meta-learning Y(w) and Y(0) directly (S-learner w/ meta-learning)meta-learned CATE model · Claims, pancytopenia, unseen drugsRATE @ 0.999: 0.21 -> 0.48Improved
zero-shot CaML versus RA-learner trained directly on test interventionsCaML · Claims, pancytopenia, unseen drugsRATE @ 0.99: 0.14 vs 0.13No clear effect
zero-shot CaML versus X-learner trained on test interventionsCaML · Claims, pancytopenia, unseen drugsRATE @ 0.999: 0.06 -> 0.48Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy