Identifiability and transportability in dynamic causal networks

Blondel, Gilles; Arias, Marta; Gavaldà, Ricard

doi:10.1007/s41060-016-0028-8

Cited by 6 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To give credit where it is due, a limited amount of causal inference work has tried to account for such feedback processes. This work includes the use of chain graphs (Lauritzen and Richardson, 2002), directed cyclic graphs (Schmidt and Murphy, 2009), the "settable systems" framework developed by econometricians Halbert White and Karim Chalak (2009), and dynamic causal networks (Blondel et al, 2017). However, as with the aforementioned causal transportability studies, these techniques have seldom been used in real applications.…”

Section: Resultsmentioning

confidence: 99%

Causal inference in travel demand modeling (and the lack thereof)

Brathwaite

Walker

2018

Journal of Choice Modelling

View full text Add to dashboard Cite

This paper is about the general disconnect that we see, both in practice and in literature, between the disciplines of travel demand modeling and causal inference. In this paper, we assert that travel demand modeling should be one of the many fields that focuses on the production of valid causal inferences, and we hypothesize about reasons for the current disconnect between the two bodies of research. Furthermore, we explore the potential benefits of uniting these two disciplines. We consider what travel demand modeling can gain from greater incorporation of techniques and perspectives from the causal inference literatures, and we briefly discuss what the causal inference literature might gain from the work of travel demand modelers. In this paper, we do not attempt to "solve" issues related to the drawing of causal inferences from travel demand models. Instead, we hope to spark a larger discussion both within and between the travel demand modeling and causal inference literatures. In particular, we hope to incite discussion about the necessity of drawing causal inferences in travel demand applications and the methods by which one might credibly do so.

show abstract

Section: Resultsmentioning

confidence: 99%

Causal inference in travel demand modeling (and the lack thereof)

Brathwaite

Walker

2018

Journal of Choice Modelling

View full text Add to dashboard Cite

show abstract

“…In well-specified causal models, a variable's CPT is the same no matter the setting of other variables and policy interventions in which it occurs; this invariant causal prediction property, along with the facts that effects depend on their direct causes and that information flows from causes to their effects over time, have been used to develop causal discovery algorithms for learning CBNs from data (Pfister, Bühlmann, & Peters, 2019). Algorithms are now well developed for using observational data both to infer the qualitative structure of a CBN, showing which variables depend on which others, and to estimate CPTs quantifying these dependencies; recent advances also allow for the possibility of unmeasured ("latent" or "hidden") causes and multiperiod changes in the distributions of variables (Blondel, Arias, & Gavaldà, 2017;Goudet et al, 2018;Jabbari et al, 2017). Conditions under which CBN models can be uniquely identified from observations and generalized to predict effects of interventions under new conditions have also been elucidated (Blondel et al, 2017).…”

Section: Learning Causal Models From Datamentioning

confidence: 99%

“…Algorithms are now well developed for using observational data both to infer the qualitative structure of a CBN, showing which variables depend on which others, and to estimate CPTs quantifying these dependencies; recent advances also allow for the possibility of unmeasured ("latent" or "hidden") causes and multiperiod changes in the distributions of variables (Blondel, Arias, & Gavaldà, 2017;Goudet et al, 2018;Jabbari et al, 2017). Conditions under which CBN models can be uniquely identified from observations and generalized to predict effects of interventions under new conditions have also been elucidated (Blondel et al, 2017). Moreover, a rich theory has been developed within the CBN framework for determining what questions can (and cannot) be answered about effects of interventions and about effects of counterfactual conditions (e.g., how many more people would have died had it not been for an intervention or policy change?)…”

Section: Learning Causal Models From Datamentioning

confidence: 99%

Answerable and Unanswerable Questions in Risk Analysis with Open‐World Novelty

Cox

2020

Risk Analysis

View full text Add to dashboard Cite

Decision analysis and risk analysis have grown up around a set of organizing questions: what might go wrong, how likely is it to do so, how bad might the consequences be, what should be done to maximize expected utility and minimize expected loss or regret, and how large are the remaining risks? In probabilistic causal models capable of representing unpredictable and novel events, probabilities for what will happen, and even what is possible, cannot necessarily be determined in advance. Standard decision and risk analysis questions become inherently unanswerable ("undecidable") for realistically complex causal systems with "open-world" uncertainties about what exists, what can happen, what other agents know, and how they will act. Recent artificial intelligence (AI) techniques enable agents (e.g., robots, drone swarms, and automatic controllers) to learn, plan, and act effectively despite open-world uncertainties in a host of practical applications, from robotics and autonomous vehicles to industrial engineering, transportation and logistics automation, and industrial process control. This article offers an AI/machine learning perspective on recent ideas for making decision and risk analysis (even) more useful. It reviews undecidability results and recent principles and methods for enabling intelligent agents to learn what works and how to complete useful tasks, adjust plans as needed, and achieve multiple goals safely and reasonably efficiently when possible, despite open-world uncertainties and unpredictable events. In the near future, these principles could contribute to the formulation and effective implementation of more effective plans and policies in business, regulation, and public policy, as well as in engineering, disaster management, and military and civil defense operations. They can extend traditional decision and risk analysis to deal more successfully with open-world novelty and unpredictable events in large-scale real-world planning, policymaking, and risk management.

show abstract

“…Very recently complete algorithms for causal identification with respect to MPDAGs have been established [Per19]. Complete algorithms are also known for dynamic causal networks, a causal analogue for dynamic Bayesian networks that evolve over time [BAG16]. Causal chain graphs (CEGs, which are similar to ADMGs) are yet another class of graphs for which identifiability of interventions has been investigated and conditions (similar to Pearl's back-door criterion) have been established [TSR10,Thw13].…”

Section: Previous Workmentioning

confidence: 99%

Learning and Sampling of Atomic Interventions from Observations

Bhattacharyya

Gayen²,

Kandasamy

et al. 2020

Preprint

View full text Add to dashboard Cite

We study the problem of efficiently estimating the effect of an intervention on a single variable (atomic interventions) using observational samples in a causal Bayesian network. Our goal is to give algorithms that are efficient in both time and sample complexity in a non-parametric setting.Tian and Pearl (AAAI '02) have exactly characterized the class of causal graphs for which causal effects of atomic interventions can be identified from observational data. We make their result quantitative. Suppose P is a causal model on a set V of n observable variables with respect to a given causal graph G with observable distribution P . Let P x denote the interventional distribution over the observables with respect to an intervention of a designated variable X with x. 1 We show that assuming that G has bounded in-degree, bounded c-components, and that the observational distribution is identifiable and satisfies certain strong positivity condition: (i) [Evaluation] There is an algorithm that outputs with probability 2/3 an evaluator for a distribution P that satisfies d TV (P x , P ) ε using m = Õ(nε −2 ) samples from P and O(mn) time. The evaluator can return in O(n) time the probability P (v) for any assignment v to V. (ii) [Generation]There is an algorithm that outputs with probability 2/3 a sampler for a distribution P that satisfies d TV (P x , P ) ε using m = Õ(nε −2 ) samples from P and O(mn) time. The sampler returns an iid sample from P with probability 1 − δ in O(nε −1 log δ −1 ) time.We extend our techniques to estimate marginals P x | Y over a given subset Y ⊆ V of variables of interest. We also show lower bounds for the sample complexity showing that our sample complexity has optimal dependence on the parameters n and ε as well as the strong positivity parameter.

show abstract

Identifiability and transportability in dynamic causal networks

Cited by 6 publications

References 22 publications

Causal inference in travel demand modeling (and the lack thereof)

Causal inference in travel demand modeling (and the lack thereof)

Answerable and Unanswerable Questions in Risk Analysis with Open‐World Novelty

Learning and Sampling of Atomic Interventions from Observations

Contact Info

Product

Resources

About