AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

Learning Treatment Representations for Downstream Instrumental Variable Regression

Shiangyi Lin, Hui Lan, Vasilis Syrgkanis

ICML 2026 · Probabilistic and scientific modeling › Treatment effects · approach: Instrumental variable estimation · 0 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
add instrument-regression term mu1 to the lambda-only regularizer (IRAE[0] to IRAE[1]) (ablation)IRAE · Quadratic data, Case 2Average test improvement: 2.2898 -> 4.6175Improved
add mu1 term (IRAE[0] to IRAE[1]) (ablation)IRAE · Quadratic data, Case 1Average test improvement: 6.1055 -> 6.4174No clear effect
add mu2 and larger bottleneck 10 (IRAE[1] to IRAE[2]) (ablation)IRAE · Quadratic data, Case 2Average test improvement: 4.6175 -> 4.5554No clear effect
add mu3 D-V dependence penalty (IRAE[2] to IRAE) (ablation)IRAE · Quadratic data, Case 2Average test improvement: 4.5554 -> 6.5253Improved
add mu3 D-V dependence penalty (IRAE[2] to IRAE) (ablation)IRAE · Quadratic data, Case 3Average test improvement: 5.2145 -> 4.9113Hurt
add mu3 D-V dependence penalty (IRAE[2] to IRAE) (ablation)IRAE · MNIST (sample size 10000, intervened alpha=1.0)Average test improvement: 0.92 -> 1.1Improved
add instrument regularization with bottleneck 10 (IRAE[1] to IRAE[2]) (ablation)IRAE · MNIST (sample size 10000, intervened alpha=1.0)Average test improvement: 0.15 -> 0.92Improved
use LIRR instead of PCALIRR · Linear data, m=50, DGP 3Average test improvement: 2.5171 -> 5.4944Improved
use LIRR instead of PCA with independent Gaussian noiseLIRR · Linear data, m=500, DGP 1Average test improvement: 1.0098 -> 1.0355No clear effect
increase dimension of X from 50 to 500LIRR · Linear data, DGP 1Average test improvement: 3.7283 -> 1.0355Hurt
use IRAE instead of vanilla AEIRAE · Quadratic data, Case 1Average test improvement: 0.4138 -> 5.7740Improved
use LIRR instead of iVAELIRR · Quadratic data, Case 2Average test improvement: 0.1192 -> 3.4711Improved

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