Timezone: »

 
Poster
Staggered Rollout Designs Enable Causal Inference Under Interference Without Network Knowledge
Mayleen Cortez · Matthew Eichhorn · Christina Yu

Thu Dec 01 02:00 PM -- 04:00 PM (PST) @ Hall J #324
in Poster Session 6 »

Randomized experiments are widely used to estimate causal effects across many domains. However, classical causal inference approaches rely on independence assumptions that are violated by network interference, when the treatment of one individual influences the outcomes of others. All existing approaches require at least approximate knowledge of the network, which may be unavailable or costly to collect. We consider the task of estimating the total treatment effect (TTE), the average difference between the outcomes when the whole population is treated versus when the whole population is untreated. By leveraging a staggered rollout design, in which treatment is incrementally given to random subsets of individuals, we derive unbiased estimators for TTE that do not rely on any prior structural knowledge of the network, as long as the network interference effects are constrained to low-degree interactions among neighbors of an individual. We derive bounds on the variance of the estimators, and we show in experiments that our estimator performs well against baselines on simulated data. Central to our theoretical contribution is a connection between staggered rollout observations and polynomial extrapolation.

Keywords: Causal Inference · staggered rollout design · network interference · polynomial interpolation · total treatment effect · global average treatment effect

Author Information

Mayleen Cortez (Cornell University)

Mayleen Cortez-Rodriguez is a third-year Applied Mathematics Ph.D. student at Cornell University. She graduated from California State University, Channel Islands in May 2020 with a B.S. in Mathematics and a minor in Computer Science. She is a National Science Foundation Graduate Research Fellowship recipient. Past research interests include mathematical modeling and reinforcement learning with applications to biology and public health. She is currently working with Dr. Christina Yu on problems in causal inference under interference.

Matthew Eichhorn (Cornell University)

I am a fourth-year PhD student in the Center for Applied Mathematics at Cornell University, where I am advised by Siddhartha Banerjee. My research focuses on the design of algorithms for combinatorial problems, particularly with applications to game theory. In addition, I have a passion for teaching and curricular development. I have helped to develop undergraduate course materials for the Cornell Math Department's Active Learning Initiative. Beyond academia, I am an avid baker and a devoted fan of the Buffalo Bills.

Christina Yu (Cornell University)

More from the Same Authors