On Evolution-Based Models for Experimentation Under Interference

2025-11-26

AI Summary

This paper addresses the estimation of population-level causal effects under network interference when the underlying network structure is unobserved. Methodologically, it introduces the “evolutionary mapping” perspective: rather than recovering the latent network, it models parallel evolutionary trajectories of outcome distributions under intervention, leveraging randomized implicit sampling to identify heterogeneous spillover effects without explicit knowledge of interference pathways. Building upon the exposure mapping axiomatization, it formulates low-dimensional recursive equations to characterize distributional evolution and integrates a causal message-passing mechanism to accommodate complex interference structures—including dense networks and influencer-driven networks. Theoretically, it establishes identifiability of causal effects under weak structural assumptions and achieves consistent estimation of population-level causal effects across diverse interference scenarios. Moreover, it precisely delineates the method’s validity boundaries in the presence of temporal trends and endogenous interference.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Causal effect estimation in networked systems is central to data-driven decision making. In such settings, interventions on one unit can spill over to others, and in complex physical or social systems, the interaction pathways driving these interference structures remain largely unobserved. We argue that for identifying population-level causal effects, it is not necessary to recover the exact network structure; instead, it suffices to characterize how those interactions contribute to the evolution of outcomes. Building on this principle, we study an evolution-based approach that investigates how outcomes change across observation rounds in response to interventions, hence compensating for missing network information. Using an exposure-mapping perspective, we give an axiomatic characterization of when the empirical distribution of outcomes follows a low-dimensional recursive equation, and identify minimal structural conditions under which such evolution mappings exist. We frame this as a distributional counterpart to difference-in-differences. Rather than assuming parallel paths for individual units, it exploits parallel evolution patterns across treatment scenarios to estimate counterfactual trajectories. A key insight is that treatment randomization plays a role beyond eliminating latent confounding; it induces an implicit sampling from hidden interference channels, enabling consistent learning about heterogeneous spillover effects. We highlight causal message passing as an instantiation of this method in dense networks while extending to more general interference structures, including influencer networks where a small set of units drives most spillovers. Finally, we discuss the limits of this approach, showing that strong temporal trends or endogenous interference can undermine identification.

Problem

• Estimating causal effects under interference in networked systems.
• Characterizing outcome evolution without exact network structure.
• Identifying conditions for consistent spillover effect estimation.

Innovation

• Evolution-based approach compensates for missing network information
• Exposure-mapping characterizes low-dimensional recursive outcome evolution
• Treatment randomization enables learning heterogeneous spillover effects implicitly