AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

On Inductive Biases for Heterogeneous Treatment Effect Estimation

Alicia Curth, Mihaela van der Schaar

NeurIPS 2021 · Probabilistic and scientific modeling › Treatment effects · approach: Meta-learners and partial identification · 11 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
Encourage shared structure across potential-outcome functions using regularizationTNet and TARNet · Simulations A and BCATE RMSEImproved
Use additive reparametrization, expressing treatment outcome as control outcome plus treatment effectTNet · Simulation A, especially with smaller treatment samplesCATE RMSEImproved
Use additive reparametrization when treated outcome is simpler than control outcomeTNet · Simulation BCATE RMSEHurt
Learn shared and private subspaces with FlexTENetTARNet · Simulations A and BCATE RMSEImproved
Remove orthogonal regularization from FlexTENet and set lambda_o=0 (ablation)FlexTENet · Simulations A and BCATE RMSEHurt
Set lambda_1 < lambda_2, regularizing CATE more heavily than potential outcomes (ablation)Soft regularization and FlexTENet · Setup A, n1=200CATE RMSEImproved
Add orthogonal regularization between shared and private subspaces (ablation)FlexTENet · Setup A, n1=200CATE RMSEImproved
Combine orthogonal regularization with lambda_1 < lambda_2 (ablation)FlexTENet · Setup A, n1=200CATE RMSEImproved
Use the flexible architecture with smaller control sample sizeFlexTENet · Setups A and B, smaller n0CATE RMSEMixed
Apply the soft regularizer to DragonNetDragonNet · Setups A and BCATE RMSEImproved
Use FlexTENet rather than SNet, with comparable shared/private feature spacesSNet · Setups A and BCATE RMSEImproved
Use reparametrization on IHDP setup C, where additive treatment effects and simpler CATE do not holdTNet · IHDP setup Cnormalized RMSE CATE: 0.32 (.01) -> 0.39 (.01) in-sample; 0.34 (.01) -> 0.40 (.01) out-of-sampleHurt

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