AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

Estimating Multi-cause Treatment Effects via Single-cause Perturbation

Zhaozhi Qian, Alicia Curth, Mihaela van der Schaar

NeurIPS 2021 · Probabilistic and scientific modeling › Treatment effects · approach: Meta-learners and partial 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
use SCP instead of NN covariate adjustmentNN · CHESS N0=500RMSE: 1.257 -> 1.117Improved
use SCP instead of NNNN · CHESS N0=1000RMSE: 1.383 -> 1.098Improved
use SCP instead of NNNN · CHESS N0=1500RMSE: 1.116 -> 1.044Improved
use SCP instead of NNNN · CHESS N0=500Ranking error: 282.3 -> 230.5Improved
use SCP instead of NNNN · CHESS N0=1000Ranking error: 321.6 -> 221.3Improved
use SCP instead of VSRVSR · CHESS N0=1500RMSE: 1.140 -> 1.044Improved
use SCP instead of DeconfounderDEC · CHESS N0=500Ranking error: 283.9 -> 230.5Improved
increase training size from N0=500 to 1500 for SCPSCP · CHESSRMSE: 1.117 -> 1.044Improved
use SCP trained on 500 vs benchmarks trained on 1500SCP · CHESSRMSE: 1.117 vs 1.116Improved
single-cause CFR / DR-CFR instead of NNNN · CHESS N0=500RMSE: 1.257 -> 2.028Hurt
remove prior knowledge of non-descendants (A-k up = empty) (ablation)SCP · Simulation K=10RMSEHurt
add augmented datasets Dk one by oneSCP · Simulation K=10RMSEImproved

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