AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

Partial Identification of Treatment Effects with Implicit Generative Models

Vahid Balazadeh Meresht, Vasilis Syrgkanis, Rahul G Krishnan

NeurIPS 2022 · Probabilistic and scientific modeling › Treatment effects · approach: Proximal and deep causal effect estimation · 2 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
optimize uniform average treatment derivative instead of ATE directly (vs GAN baseline)generative partial-identification model · nonlinear back-door, IV, leaky mediationATE bound tightness and stabilityImproved
use fixed basis-function moment-matching method instead of oursmoment-matching baseline · nonlinear IVbound validityHurt
increase sample size N from 500 to 5000ATD bounds · front-door, back-door, IV, leaky mediationATD bound widthImproved
increase measurement noise level on ACIC covariatesATE bounds · ACIC 2019bound informativenessHurt
increase standard deviation of the Gaussian tail density mu-tilde (ablation)UATD objective · linear process Y = T + N(0,1)response curve smoothnessImproved
apply method to binary front-door graphours · Front-door (Discrete)ATE bound: (0.4374, 0.5322) vs true 0.5085Improved
apply method to binary IV graphours · IV (Discrete)ATE bound vs optimal: (-0.5629, -0.0821) vs (-0.55, -0.15)Mixed

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