Comment: Graphical Models, Causality and Intervention

By Judea Pearl. 2016

137 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Domain Adaptation by Using Causal Inference to Predict Invariant Conditional Distributions (2017, background): “…domain, but much more accurate ones in the target domain. across the different domains, allows one to transfer knowledge from one domain to the other in order to address the problem of domain adaptation [Spirtes et al., 2000, Storkey, 2009, Schölkopf et al., 2012, Bareinboim and Pearl, 2016].”
Theoretical Impediments to Machine Learning With Seven Sparks from the Causal Revolution (2018, background): “It can be used both for re-adjusting learned policies to circumvent environmental changes and for controlling bias due to non-representative samples (Bareinboim and Pearl, 2016).”
Comment: Understanding Simpson’s Paradox (2014, unknown): “It involves a simple graphical condition called “back-door” (Pearl 1993) which traces paths in the causal diagram and assures that all spurious paths from treatment to outcome are intercepted by the third variable.”
CAUSAL INTERPRETATIONS OF BLACK-BOX MODELS (2019, unknown): “Interestingly, Equation (2) that deﬁnes PDP is exactly the same as the famous back-door adjustment formula of Pearl (1993) to identify causal effect of X S on Y from observational data.”
Interpretation and identification of causal mediation. (2014, unknown): “The do-operator was devised to make this distinction formal (Pearl, 1993).”
A Complete Generalized Adjustment Criterion (2015, background): “Our results entail several existing, less general or less powerful ones (Pearl, 1993; Shpitser et al., 2010; Textor and Lískiewicz, 2011; van der Zander et al., 2014; Maathuis and Colombo, 2015) as special cases.”
On the Identification of Causal Effects (2015, background): “…absence of unmeasured confounders, all causal effects are identifiable, that is, the joint response of any set Y of variables to intervention on a set T of action variables, denoted Pt(y), 1 can be estimated consistently from nonexperimental data [Robins, 1986, Spirtes et al., 1993, Pearl, 1993].”
The seven tools of causal inference, with reflections on machine learning (2019, methodology): “” In particular, the task of selecting an appropriate set of covariates to control for confounding has been reduced to a simple “roadblocks” puzzle manageable by a simple algorithm [Pearl 1993].”
Graphical criteria for efficient total effect estimation via adjustment in causal linear models (2019, background): “The best-known such criterion is probably the back-door criterion (Pearl, 1993), which is sufficient for adjustment.”
TRYGVE HAAVELMO AND THE EMERGENCE OF CAUSAL CALCULUS (2014, unknown): “Haavelmo’s approach was later transformed by Strotz and Wold (1960) into the operation of “wiping out” the equation altogether, and was further translated into graphical models as “wiping out” incoming arrows into the manipulated variable (Pearl, 1993; Spirtes et al., 1993).”
Causes of Effects and Effects of Causes (2015, background): “…(4) thus provides a more refined measure of causation, which can be used for monotonic Yx(u) whenever the causal effect Pðy j doðxÞÞ can be estimated, from either randomized trials or graph-assisted observational studies (e.g., through the back-door criterion, Pearl 1993, or the do-calculus).”
Causal Transfer Learning (2017, background): “These interventions are not limited to the perfect (“surgical”) interventions modeled by the do-operator of Pearl [2009] (or alternatively, by “force variables” Pearl [1993]), but can also model more general types of interventions such as mechanism changes.”
Causal inference and the data-fusion problem (2016, methodology): “The following criterion, named “backdoor” (17), provides a graphical method of selecting such a set of factors for adjustment.”
Cost overruns in transportation infrastructure projects: Sowing the seeds for a probabilistic theory of causation (2016, unknown): “To understand the nomological context of cost overruns that may prevail, it is suggested that ‘sensemaking’ (e.g., Wieck et al , 1988, 1993; 1995; Goh et al ., 2012) form the epistemological underpinning used to derive causal inferences.”
Causal and Counterfactual Inference (2018, methodology): “Conditions for relaxing restriction 1 are given in (Pearl, 2009b, p. 338)Shpitser et al. (2010); Pearl and Paz (2014)(7)).”
Joint Causal Inference from Multiple Contexts (2016, background): “Modeling a perfect intervention in this way is similar to the concept of “force variables” introduced by Pearl (1993).”