# Experimenting under Stochastic Congestion

arXiv: https://arxiv.org/abs/2302.12093 · HTML: https://mlsanity.com/paper/2302.12093/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Feb 2023.

## Built on by (most cited first, 6 of 6)

- [Data-Driven Switchback Experiments: Theoretical Tradeoffs and Empirical Bayes Designs](https://mlsanity.com/paper/2406.06768/) (2024)
- [Double Machine Learning for Causal Inference under Shared-State Interference](https://mlsanity.com/paper/2504.08836/) (2025)
- [Choosing the Better Bandit Algorithm under Data Sharing: When Do A/B Experiments Work?](https://mlsanity.com/paper/2507.11891/) (2025)
- [Tackling Interference Induced by Data Training Loops in A/B Tests: A Weighted Training Approach](https://mlsanity.com/paper/2310.17496/) (2023)
- [Clustered Switchback Designs for Experimentation Under Spatio-temporal Interference](https://mlsanity.com/paper/2312.15574/) (2023)
- [On Evolution-Based Models for Experimentation Under Interference](https://mlsanity.com/paper/2511.21675/) (2025)

## Abstract (arXiv)

We study randomized experiments in a service system when stochastic congestion can arise from temporarily limited supply or excess demand. Such congestion gives rise to cross-unit interference between the waiting customers, and analytic strategies that do not account for this interference may be biased. In current practice, one of the most widely used ways to address stochastic congestion is to use switchback experiments that alternatively turn a target intervention on and off for the whole system. We find, however, that under a queueing model for stochastic congestion, the standard way of analyzing switchbacks is inefficient, and that estimators that leverage the queueing model can be materially more accurate. Additionally, we show how the queueing model enables estimation of total policy gradients from unit-level randomized experiments, thus giving practitioners an alternative experimental approach they can use without needing to pre-commit to a fixed switchback length before data collection.