Causal Discovery in High-Dimensional Point Process Networks with Hidden Nodes

Abstract

Thanks to technological advances leading to near-continuous time observations, emerging multivariate point process data offer new opportunities for causal discovery. However, a key obstacle in achieving this goal is that many relevant processes may not be observed in practice. Naïve estimation approaches that ignore these hidden variables can generate misleading results because of the unadjusted confounding. To plug this gap, we propose a deconfounding procedure to estimate high-dimensional point process networks with only a subset of the nodes being observed. Our method allows flexible connections between the observed and unobserved processes. It also allows the number of unobserved processes to be unknown and potentially larger than the number of observed nodes. Theoretical analyses and numerical studies highlight the advantages of the proposed method in identifying causal interactions among the observed processes.

Stimulus type shapes the topology of cellular functional networks in mouse visual cortex

“…Hence, the potential incorporation of indirect interactions within functional connections enables the examination of effective interaction between neurons, even with unrecorded units. Third, despite the demonstrated efficacy of various methods to correct for latent variables in the construction of causal graphs 107 – 110 , the applicability of these models in real-world analyses, particularly in the absence of ground truth connectivity, remains uncertain. It is imperative to underscore that, our study primarily focuses on functional connectivity, which reveals correlations instead of causality between neural activities under different stimulus conditions 111 .…”

Causal Inference for Heterogeneous Data and Information Theory

“…Temporal causal models. Papers [ 8 , 9 ] consider causal models using time. Paper [ 8 ] investigates causal discovery in high-dimensional point process networks with hidden nodes.…”

“…Papers [ 8 , 9 ] consider causal models using time. Paper [ 8 ] investigates causal discovery in high-dimensional point process networks with hidden nodes. A big challenge in the multivariate causal discovery is the confounding problem.…”

“…A big challenge in the multivariate causal discovery is the confounding problem. Paper [ 8 ] proposes a deconfounding procedure to estimate high-dimensional point process networks with only a subset of the nodes being observed. The method allows flexible connections between the observed and unobserved processes.…”