• Corpus ID: 237260220

Causal Expectation-Maximisation

@inproceedings{Zaffalon2020CausalE,
  title={Causal Expectation-Maximisation},
  author={Marco Zaffalon and Alessandro Antonucci and Rafael Caba{\~n}as},
  year={2020}
}
Structural causal models are the basic modelling unit in Pearl’s causal theory; in principle they allow us to solve counterfactuals, which are at the top rung of the ladder of causation. But they often contain latent variables that limit their application to special settings. This appears to be a consequence of the fact, proven in this paper, that causal inference is NP-hard even in models characterised by polytree-shaped graphs. To deal with such a hardness, we introduce the causal EM… 

Figures and Tables from this paper

Partial Counterfactual Identification from Observational and Experimental Data
TLDR
It is shown that all counterfactual distributions in an arbitrary structural causal model (SCM) could be generated by a canonical family of SCMs with the same causal diagram where unobserved (exogenous) variables are discrete with a finite domain.

References

SHOWING 1-10 OF 17 REFERENCES
Structural Causal Models Are (Solvable by) Credal Networks
TLDR
This work shows that endogenous observations induce linear constraints on the probabilities of the exogenous variables, which allows to exactly map a causal model into a credal network and hence to systematically compute causal inferences.
An Automated Approach to Causal Inference in Discrete Settings
TLDR
This work presents a general and automated approach to causal inference in discrete settings, and presents an algorithm to automatically bound causal effects using efficient dual relaxation and spatial branch-and-bound techniques.
Studies in causal reasoning and learning
TLDR
It is shown how useful information on the probabilities of causation can be extracted from empirical data, and how data from both experimental and nonexperimental studies can be combined to yield information that neither study alone can provide.
On Pearl’s Hierarchy and the Foundations of Causal Inference
TLDR
This chapter develops a novel and comprehensive treatment of the Pearl Causal Hierarchy through two complementary lenses: one logical-probabilistic and another inferential-graphical, and investigates an inferential system known as do-calculus, showing how it can be suf­ ficient, and in many cases necessary, to allow inferences across the PCH’s layers.
Symbolic Computation of Tight Causal Bounds
TLDR
A general approach for computation of bounds is developed, proving that if the problem can be stated as a linear program, then the true global extrema result in tight bounds.
Causality in Bayesian Belief Networks
Probabilities Of Causation: Three Counterfactual Interpretations And Their Identification
TLDR
It is shown thatnecessity and sufficiency are two independent aspects of causation, and that both should be invoked in the construction of causal explanations for specific scenarios.
CREDICI: A Java Library for Causal Inference by Credal Networks
TLDR
Credal networks are an extension of Bayesian networks where local probability mass functions are only constrained to belong to given, so-called credal, sets and CREDICI is based on the recent work of Zaffalon et al. (2020), where an equivalence between Pearl's structural causal models and credal networks has been derived.
What Counterfactuals Can Be Tested
TLDR
This paper provides a complete characterization of "testable counterfactuals," namely,Counterfactual statements whose probabilities can be inferred from physical experiments, and provides complete procedures for discerning whether a givencounterfactual is testable and, if so, expressing its probability in terms of experimental data.
Probabilistic Graphical Models - Principles and Techniques
TLDR
The framework of probabilistic graphical models, presented in this book, provides a general approach for causal reasoning and decision making under uncertainty, allowing interpretable models to be constructed and then manipulated by reasoning algorithms.
...
1
2
...