Causal Discovery in the Presence of Distribution Shifts
It is commonplace to encounter heterogeneous or nonstationary data, of which the underlying generating process changes across domains or over time. Such a distribution shift feature presents both challenges and opportunities for causal discovery, as well as for various ML tasks.
Causal Discovery from Nonstationary/Heterogeneous Data
In this work, we develop a framework for causal discovery from such data that can efficiently locate variables with changing causal mechanisms, reliably recover the causal structure, and extract a low-dimensional representation of changes to visualize how the causal mechanism changes over time. Moreover, by making use of the independent change property between causal modules, with invariance as a special case, we make it explicit and precise how distribution shifts benefit causal discovery. [paper][ code-Matlab ] [ code-Python ]
Federated Causal Discovery from Heterogeneous Data
Conventional causal discovery methods rely on centralized data, which is inconsistent with the decentralized nature of data in many real-world situations. This discrepancy has motivated the development of federated causal discovery (FCD) approaches. However, existing FCD methods may be limited by their potentially restrictive assumptions of identifiable functional causal models or homogeneous data distributions, narrowing their applicability in diverse scenarios. In this paper, we propose a novel FCD method attempting to accommodate arbitrary causal models and heterogeneous data. We first utilize a surrogate variable corresponding to the client index to account for the data heterogeneity across different clients. We then develop a federated conditional independence test (FCIT) for causal skeleton discovery and establish a federated independent change principle (FICP) to determine causal directions. These approaches involve constructing summary statistics as a proxy of the raw data to protect data privacy. Owing to the nonparametric properties, FCIT and FICP make no assumption about particular functional forms, thereby facilitating the handling of arbitrary causal models. [pdf]
Causal Discovery and Forecasting in Nonstationary Environments with State-Space Models
Given nonstationary processes where the causal relations may change over time, how can we discover the time-varying causal relationships, and meanwhile predict future values of variables of interest in a principled way? In this work, we show that causal discovery and forecasting for nonstationary processes can be put under the same umbrella. Particularly, by exploiting a particular type of state-space model that represents the processes, we find that nonstationarity actually helps to identify causal structure and that forecasting naturally benefits from the learned causal representation. Moreover, given the causal model, we can directly treat forecasting as a problem in Bayesian inference that exploits the time-varying property of the data and adapts to new observations in a principled manner. [paper] [ code ][poster]
Specific and Shared Causal Relation Modeling and Mechanism-Based Clustering
In this work, we develop a unified framework for causal discovery and mechanism-based group identification. In particular, we propose a specific and shared causal model (SSCM), which takes into account the variabilities of causal relations across individuals/groups and leverages their commonalities to achieve statistically reliable estimation. The learned SSCM gives the specific causal knowledge for each individual as well as the general trend over the population. Moreover, the estimated model directly provides the group information of each individual. [paper] [ code ][poster]
Multi-Domain Causal Structure Learning in Linear Systems
In this work, we study the problem of causal structure learning in linear systems from observational data given in multiple domains, across which the causal coefficients and/or the distribution of the exogenous noises may vary. The main tool used in our approach is the principle that in a causally sufficient system, the causal modules, as well as their included parameters, change independently across domains. [paper]
Identification of Time-Dependent Causal Model: A Gaussian Process Treatment
In this work, we present a novel approach to modeling time-dependent causal influences. We show that by introducing time information as a common cause for the observed processes, we can model the time-varying causal influences between the observed processes, as well as the influence from a certain type of unobserved confounders. We propose a principled way for the estimation by extending Gaussian Process regression, which enables an automatic way to learn how the causal model changes over time. [paper][ code ][poster]
Biwei Huang 