Spatio-Temporal Graphical Counterfactuals: An Overview

Abstract

Counterfactual thinking is a crucial yet challenging topic for artificial intelligence to learn knowledge from data and ultimately improve performance for new scenarios. Many research works, including the Potential Outcome Model (POM) and the Structural Causal Model (SCM), have been proposed to address this. However, their modeling, theoretical foundations, and application approaches often differ. Moreover, there is a lack of graphical approaches for inferring spatio-temporal counterfactuals, that account for spatial and temporal interactions among multiple units. Thus, in this work, we aim to present a survey that compares and discusses different counterfactual models, theories and approaches. Additionally, we propose a unified graphical causal framework to infer spatio-temporal counterfactuals.

Published

References in corpus (13)

• Synchronization in complex networks
• Robustness of a Network of Networks
• Scalable Tensor Factorizations for Incomplete Data
• The fundamental advantages of temporal networks
• Explosive transitions in complex networks' structure and dynamics: percolation and synchronization
• Learning high-dimensional directed acyclic graphs with latent and selection variables
• Network resilience
• Causal Network Inference by Optimal Causation Entropy
• The synchronized dynamics of time-varying networks
• The low-rank hypothesis of complex systems
• Autonomous inference of complex network dynamics from incomplete and noisy data
• Foundations of Structural Causal Models with Cycles and Latent Variables
• Identifying Unique Spatial-Temporal Bayesian Network without Markov Equivalence