On the Asymptotics of Graph Cut Objectives for Experimental Designs of Network A/B Testing

TL;DR

This paper explores the asymptotic behavior of graph cut objectives in network-based A/B testing, providing theoretical insights to improve experimental design in social networks.

Contribution

It develops asymptotic distributions of graph cut objectives, linking network design criteria with statistical properties for better A/B test planning.

Findings

Derived asymptotic distributions for graph cut objectives.

Enabled rerandomization algorithms for network A/B testing.

Linked design criteria with statistical properties in social networks.

Abstract

A/B testing is an effective way to assess the potential impacts of two treatments. For A/B tests conducted by IT companies, the test users of A/B testing are often connected and form a social network. The responses of A/B testing can be related to the network connection of test users. This paper discusses the relationship between the design criteria of network A/B testing and graph cut objectives. We develop asymptotic distributions of graph cut objectives to enable rerandomization algorithms for the design of network A/B testing under two scenarios.