Timezone: »

 
Spotlight
Quantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden Confounding
Andrew Jesson · Sören Mindermann · Yarin Gal · Uri Shalit

Wed Jul 21 07:30 PM -- 07:35 PM (PDT) @ None

We study the problem of learning conditional average treatment effects (CATE) from high-dimensional, observational data with unobserved confounders. Unobserved confounders introduce ignorance---a level of unidentifiability---about an individual's response to treatment by inducing bias in CATE estimates. We present a new parametric interval estimator suited for high-dimensional data, that estimates a range of possible CATE values when given a predefined bound on the level of hidden confounding. Further, previous interval estimators do not account for ignorance about the CATE associated with samples that may be underrepresented in the original study, or samples that violate the overlap assumption. Our interval estimator also incorporates model uncertainty so that practitioners can be made aware of such out-of-distribution data. We prove that our estimator converges to tight bounds on CATE when there may be unobserved confounding and assess it using semi-synthetic, high-dimensional datasets.

Author Information

Andrew Jesson (University of Oxford)
Sören Mindermann (University of Oxford)

I'm a research intern at the Center for Human-Compatible AI (CHAI) group at UC Berkeley, working on active Inverse Reward Design. Previously I worked in Oxford on a theoretical paper on inverse RL for bounded agents and did my MSc with Peter Dayan at UCL.

Yarin Gal (University of Oxford)
Uri Shalit (Technion)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors