mlsanityData to 14 Oct 2026

A/B testing under Interference with Partial Network Information

arXiv 2404.10547 ↗First posted 16 Apr 2024cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

A/B tests are often required to be conducted on subjects that might have social connections. For e.g., experiments on social media, or medical and social interventions to control the spread of an epidemic. In such settings, the SUTVA assumption for randomized-controlled trials is violated due to network interference, or spill-over effects, as treatments to group A can potentially also affect the control group B. When the underlying social network is known exactly, prior works have demonstrated how to conduct A/B tests adequately to estimate the global average treatment effect (GATE). However, in practice, it is often impossible to obtain knowledge about the exact underlying network. In this paper, we present UNITE: a novel estimator that relax this assumption and can identify GATE while only relying on knowledge of the superset of neighbors for any subject in the graph. Through theoretical analysis and extensive experiments, we show that the proposed approach performs better in comparison to standard estimators.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 29 of its 68 references that are in the map. First posted on arXiv 16 Apr 2024.

Builds on

29 of 68 references.

References · 29

  1. 2016Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learningcited by 286
  2. 2018The Blessings of Multiple Causescited by 104
  3. 2013Graph cluster randomization: network exposure to multiple universescited by 42
  4. 2014Design and analysis of experiments in networks: Reducing bias from interferencecited by 42
  5. 2018Bounds on the conditional and average treatment effect with unobserved confounding factorscited by 39
  6. 2017Causal inference for social network datacited by 31
  7. 2019Causal Inference Under Interference And Network Uncertaintycited by 27
  8. 2020Sense and Sensitivity Analysis: Simple Post-Hoc Analysis of Bias Due to Unobserved Confoundingcited by 25
  9. 2021Interference, Bias, and Variance in Two-Sided Marketplace Experimentation: Guidance for Platformscited by 22
  10. 2019On Multi-Cause Causal Inference with Unobserved Confounding: Counterexamples, Impossibility, and Alternativescited by 20
  11. 2022Estimating Total Treatment Effect in Randomized Experiments with Unknown Network Structurecited by 18
  12. 2020Identifying effects of multiple treatments in the presence of unmeasured confoundingcited by 16
17 more
  1. 2018Multiple Causal Inference with Latent Confoundingcited by 14
  2. 2020Causal Inference with Spatio-temporal Data: Estimating the Effects of Airstrikes on Insurgent Violence in Iraqcited by 14
  3. 2020Causal Network Motifs: Identifying Heterogeneous Spillover Effects in A/B Testscited by 14
  4. 2015Nonparametric Bounds and Sensitivity Analysis of Treatment Effectscited by 12
  5. 2017Elements of estimation theory for causal effects in the presence of network interferencecited by 12
  6. 2020Heterogeneous Treatment and Spillover Effects under Clustered Network Interferencecited by 11
  7. 2015Private Causal Inferencecited by 10
  8. 2018Regression adjustments for estimating the global treatment effect in experiments with interferencecited by 10
  9. 2015Model-assisted design of experiments in the presence of network correlated outcomescited by 9
  10. 2020Experimental Design under Network Interferencecited by 9
  11. 2017Testing for arbitrary interference on experimentation platformscited by 8
  12. 2021The Local Approach to Causal Inference under Network Interferencecited by 6
  13. 2022Cluster Randomized Designs for One-Sided Bipartite Experimentscited by 6
  14. 2016Importance Sampling with Unequal Supportcited by 5
  15. 2021Causal Inference under Network Interference with Noisecited by 4
  16. 2020Computational Causal Inferencecited by 3
  17. 2022Privacy Aware Experiments without Cookiescited by 1

Built on by

Not cited yet.