# Journey to the Centre of Cluster: Harnessing Interior Nodes for A/B Testing under Network Interference

arXiv: https://arxiv.org/abs/2602.04457 · HTML: https://mlsanity.com/paper/2602.04457/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 38 references that are in the map.
- First posted on arXiv 4 Feb 2026.

## Builds on (most cited first, 22 of 22)

- [Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering](https://mlsanity.com/paper/1606.09375/) (2016)
- [Prediction-Powered Inference](https://mlsanity.com/paper/2301.09633/) (2023)
- [PPI++: Efficient Prediction-Powered Inference](https://mlsanity.com/paper/2311.01453/) (2023)
- [Identification and estimation of treatment and interference effects in observational studies on networks](https://mlsanity.com/paper/1609.06245/) (2016)
- [Graph cluster randomization: network exposure to multiple universes](https://mlsanity.com/paper/1305.6979/) (2013)
- [Design and analysis of experiments in networks: Reducing bias from interference](https://mlsanity.com/paper/1404.7530/) (2014)
- [Causal Inference Under Interference And Network Uncertainty](https://mlsanity.com/paper/1907.00221/) (2019)
- [Estimating Total Treatment Effect in Randomized Experiments with Unknown Network Structure](https://mlsanity.com/paper/2205.12803/) (2022)
- [Randomized Graph Cluster Randomization](https://mlsanity.com/paper/2009.02297/) (2020)
- [Causal clustering: design of cluster experiments under network interference](https://mlsanity.com/paper/2310.14983/) (2023)
- [Model-assisted design of experiments in the presence of network correlated outcomes](https://mlsanity.com/paper/1507.00803/) (2015)
- [Design and Analysis of Bipartite Experiments under a Linear Exposure-Response Model](https://mlsanity.com/paper/2103.06392/) (2021)
- [Staggered Rollout Designs Enable Causal Inference Under Interference Without Network Knowledge](https://mlsanity.com/paper/2205.14552/) (2022)
- [Graph Neural Networks for Causal Inference Under Network Confounding](https://mlsanity.com/paper/2211.07823/) (2022)
- [Cluster-Adaptive Network A/B Testing: From Randomization to Estimation](https://mlsanity.com/paper/2008.08648/) (2020)
- [Cluster Randomized Designs for One-Sided Bipartite Experiments](https://mlsanity.com/paper/2210.16415/) (2022)
- [A systematic investigation of classical causal inference strategies under mis-specification due to network interference](https://mlsanity.com/paper/1810.08259/) (2018)
- [Higher-Order Causal Message Passing for Experimentation with Complex Interference](https://mlsanity.com/paper/2411.00945/) (2024)
- [Optimized Covariance Design for AB Test on Social Network under Interference](https://mlsanity.com/paper/2311.14042/) (2023)
- [Independent-Set Design of Experiments for Estimating Treatment and Spillover Effects under Network Interference](https://mlsanity.com/paper/2312.04026/) (2023)
- [The Conflict Graph Design: Estimating Causal Effects under Arbitrary Neighborhood Interference](https://mlsanity.com/paper/2411.10908/) (2024)
- [Just Ramp-up: Unleash the Potential of Regression-based Estimator for A/B Tests under Network Interference](https://mlsanity.com/paper/2410.12740/) (2024)

## Abstract (arXiv)

A/B testing on platforms often faces challenges from network interference, where a unit's outcome depends not only on its own treatment but also on the treatments of its network neighbors. To address this, cluster-level randomization has become standard, enabling the use of network-aware estimators. These estimators typically trim the data to retain only a subset of informative units, achieving low bias under suitable conditions but often suffering from high variance. In this paper, we first demonstrate that the interior nodes - units whose neighbors all lie within the same cluster - constitute the vast majority of the post-trimming subpopulation. In light of this, we propose directly averaging over the interior nodes to construct the mean-in-interior (MII) estimator, which circumvents the delicate reweighting required by existing network-aware estimators and substantially reduces variance in classical settings. However, we show that interior nodes are often not representative of the full population, particularly in terms of network-dependent covariates, leading to notable bias. We then augment the MII estimator with a counterfactual predictor trained on the entire network, allowing us to adjust for covariate distribution shifts between the interior nodes and full population. By rearranging the expression, we reveal that our augmented MII estimator embodies an analytical form of the point estimator within prediction-powered inference framework. This insight motivates a semi-supervised lens, wherein interior nodes are treated as labeled data subject to selection bias. Extensive and challenging simulation studies demonstrate the outstanding performance of our augmented MII estimator across various settings.