CF-GODE: Continuous-Time Causal Inference for Multi-Agent Dynamical Systems

Jiang, Song; Huang, Z; Luo, Xiao; Sun, Yizhou

doi:10.1145/3580305.3599272

Cited by 1 publication

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To tackle these challenges, we propose a novel causal inference framework: the Causal Graph Ordinary Differential Equations (CAG-ODE) to estimate the continuous counterfactual outcome of a multi-agent dynamical system in the presence of multiple treatments and time-varying confounding and interference. Building upon the recent success of graph ordinary differential equations (ODE) in capturing the continuous interaction among agents [13,14,17,27], our key innovation is to learn time-dependent representations of simultaneous treatments and incorporate them into the ODE function to accurately account for their casual effects on the system. As nodes and edges are jointly evolving, we utilize two coupled treatment-induced ODE functions to account for their respective dynamics.…”

Section: Introductionmentioning

confidence: 99%

Causal Graph ODE: Continuous Treatment Effect Modeling in Multi-agent Dynamical Systems

Huang,

Hwang,

Zhang

et al. 2024

Proceedings of the ACM Web Conference 2024

View full text Add to dashboard Cite

Real-world multi-agent systems are often dynamic and continuous, where the agents co-evolve and undergo changes in their trajectories and interactions over time. For example, the COVID-19 transmission in the U.S. can be viewed as a multi-agent system, where states act as agents and daily population movements between them are interactions. Estimating the counterfactual outcomes in such systems enables accurate future predictions and effective decisionmaking, such as formulating COVID-19 policies. However, existing methods fail to model the continuous dynamic effects of treatments on the outcome, especially when multiple treatments (e.g., "stayat-home" and "get-vaccine" policies) are applied simultaneously. To tackle this challenge, we propose Causal Graph Ordinary Differential Equations (CAG-ODE), a novel model that captures the continuous interaction among agents using a Graph Neural Network (GNN) as the ODE function. The key innovation of our model is to learn time-dependent representations of treatments and incorporate them into the ODE function, enabling precise predictions of potential outcomes. To mitigate confounding bias, we further propose two domain adversarial learning-based objectives, which enable our model to learn balanced continuous representations that are not affected by treatments or interference. Experiments on * Equally contributed.

show abstract

Section: Introductionmentioning

confidence: 99%