11,930 Hits in 4.1 sec

Strong Faithfulness and Uniform Consistency in Causal Inference [article]

Jiji Zhang, Peter L. Spirtes
2012 arXiv   pre-print
In the sense of pointwise consistency, several reliable causal inference algorithms have been established under the Markov and Faithfulness assumptions [Pearl 2000, Spirtes et al. 2001].  ...  In the sense of uniform consistency, however, reliable causal inference is impossible under the two assumptions when time order is unknown and/or latent confounders are present [Robins et al. 2000].  ...  Acknowledgement We thank Clark Glymour and Tianjiao Chu for helpful discussions, and the referees for valuable comments.  ... 
arXiv:1212.2506v1 fatcat:aqnqzamehvh4nlmm5nhdaqa6ba

Geometry of the faithfulness assumption in causal inference

Caroline Uhler, Garvesh Raskutti, Peter Bühlmann, Bin Yu
2013 Annals of Statistics  
Many algorithms for inferring causality rely heavily on the faithfulness assumption.  ...  However, due to sampling error the faithfulness condition alone is not sufficient for statistical estimation, and strong-faithfulness has been proposed and assumed to achieve uniform or high-dimensional  ...  Most of the work by the first and second author was carried out while being at ETH Zürich and UC Berkeley, respectively.  ... 
doi:10.1214/12-aos1080 fatcat:2o54ywsujzdpfa3l2e4qpjb5ti

A Uniformly Consistent Estimator of non-Gaussian Causal Effects Under the k-Triangle-Faithfulness Assumption [article]

Shuyan Wang, Peter Spirtes
2021 arXiv   pre-print
The k-Triangle-Faithfulness Assumption is a strictly weaker assumption that avoids some implausible implications of the Strong Causal Faithfulness Assumption and also allows for uniformly consistent estimates  ...  Kalisch and Bühlmann (2007) showed that for linear Gaussian models, under the Causal Markov Assumption, the Strong Causal Faithfulness Assumption, and the assumption of causal sufficiency, the PC algorithm  ...  However, the Strong Causal Faithfulness Assumption can be weakened to the strictly weaker (for some values of k) k-Faithfulness Assumption while still achieving uniform consistency.  ... 
arXiv:2107.01333v2 fatcat:lkztmto35jhyrd3yx4zv6rnnka

Detection of Unfaithfulness and Robust Causal Inference

Jiji Zhang, Peter Spirtes
2008 Minds and Machines  
The latter, we argue, is related to the possibility of controlling the probability of large errors with finite sample size ("uniform consistency") in causal inference.  ...  Many algorithms proposed in the machine learning community for inferring causality from data are grounded on two assumptions, known as the Causal Markov Condition and the Causal Faithfulness Condition.  ...  the definition of uniform consistency.  ... 
doi:10.1007/s11023-008-9096-4 fatcat:qdb4zqkwebfr3bxocninod73ly

A Uniformly Consistent Estimator of Causal Effects under the $k$-Triangle-Faithfulness Assumption

Peter Spirtes, Jiji Zhang
2014 Statistical Science  
However, the Strong Faithfulness assumption may be false with high probability in many domains.  ...  Res. 8 (2007) 613-636] described a uniformly consistent estimator of the Markov equivalence class of a linear Gaussian causal structure under the Causal Markov and Strong Causal Faithfulness assumptions  ...  In section 4, we examine weakening the Strong Causal Faithfulness Assumption and modification of the estimation procedures that preserves uniform consistency.  ... 
doi:10.1214/13-sts429 fatcat:yz5mc74smrgnffo22lfpzizy3m

Introduction to the Epistemology of Causation

Frederick Eberhardt
2009 Philosophy Compass  
They belong to a large array of possible assumptions and conditions about causal relations, whose various combinations limit the possibilities of acquiring causal knowledge in different ways.  ...  How much and in what detail the causal structure can be discovered from what kinds of data depends on the particular set of assumptions one is able to make.  ...  Together, causal Markov and faithfulness enable the inference from independence and dependence constraints in a probability distribution to features of the underlying causal structure, even though they  ... 
doi:10.1111/j.1747-9991.2009.00243.x fatcat:izvut73hw5f2dgmcrdj54hruia

Learning directed acyclic graphs based on sparsest permutations [article]

Garvesh Raskutti, Caroline Uhler
2019 arXiv   pre-print
For constraint-based methods, statistical consistency guarantees typically rely on the faithfulness assumption, which has been show to be restrictive especially for graphs with cycles in the skeleton.  ...  However, there is only limited work on consistency guarantees for score-based and hybrid algorithms and it has been unclear whether consistency guarantees can be proven under weaker conditions than the  ...  Acknowledgements We wish to thank Peter Bühlmann for various valuable discussions and insights. We also thank Peter Spirtes for helpful comments regarding the single-path-faithfulness assumption.  ... 
arXiv:1307.0366v4 fatcat:eopveag3dfcx7epwykjrwcf5qq

Learning directed acyclic graph models based on sparsest permutations

Garvesh Raskutti, Caroline Uhler
2018 Stat  
Acknowledgements We wish to thank Peter Bühlmann for various valuable discussions and insights. We also thank Peter Spirtes for helpful comments regarding the single-path-faithfulness assumption.  ...  GR was partially supported by the National Science Foundation under Grant DMS-1127914 to the Statistical and Applied Mathematical Sciences Institute.  ...  We proved uniform consistency of the SP algorithm under the λ-strong SMR assumption, which is strictly weaker than the λ-strong faithfulness assumption introduced in [26] .  ... 
doi:10.1002/sta4.183 fatcat:rl3jsylabff6zpih2ppjipazbe

An Empirical Study of the Simplest Causal Prediction Algorithm

Jerome Cremers, Joris M. Mooij
2015 Conference on Uncertainty in Artificial Intelligence  
Our findings illustrate that violations of strong faithfulness become increasingly likely in the presence of many latent variables, and this can significantly deterioriate the accuracy of constraint-based  ...  causal prediction algorithms that assume faithfulness.  ...  We thank Oliver Stegle, Barbara Rakitsch, Tom Claassen and Ioannis Tsamardinos for helpful discussions. We thank Philip Versteeg for proofreading the manuscript.  ... 
dblp:conf/uai/CremersM15 fatcat:hrptozsjsnhppiyzhr6d6r266q

Robust Causal Estimation in the Large-Sample Limit without Strict Faithfulness [article]

Ioan Gabriel Bucur, Tom Claassen, Tom Heskes
2017 arXiv   pre-print
We introduce an alternative approach by replacing strict faithfulness with a prior that reflects the existence of many 'weak' (irrelevant) and 'strong' interactions.  ...  Both rely on faithfulness to infer that the latter only influences the target effect via the cause variable.  ...  TC was supported by NWO grant 612.001.202 (MoCoCaDi), and EU-FP7 grant n.603016 (MATRICS).  ... 
arXiv:1704.01864v1 fatcat:busluonkmfhj7l64igcr7ktxpa

Causal Confusion in Imitation Learning [article]

Pim de Haan, Dinesh Jayaraman, Sergey Levine
2019 arXiv   pre-print
We show that causal misidentification occurs in several benchmark control domains as well as realistic driving settings, and validate our solution against DAgger and other baselines and ablations.  ...  We point out that ignoring causality is particularly damaging because of the distributional shift in imitation learning.  ...  Acknowledgments: We would like to thank Karthikeyan Shanmugam and Shane Gu for pointers to prior work early in the project, and Yang Gao, Abhishek Gupta, Marvin Zhang, Alyosha Efros, and Roberto Calandra  ... 
arXiv:1905.11979v2 fatcat:4keinsnl3jaxli5girymqlrye4

Identification of causal relations in neuroimaging data with latent confounders: An instrumental variable approach

Moritz Grosse-Wentrup, Dominik Janzing, Markus Siegel, Bernhard Schölkopf
2016 NeuroImage  
We consider the task of inferring causal relations in brain imaging data with latent confounders.  ...  neural processes, even in the presence of latent confounders.  ...  Causal inference in stimulus-based paradigms Let s, x, and y be three random variables with a joint probability distribution p(s, x, y) that is faithful to its generating DAG.  ... 
doi:10.1016/j.neuroimage.2015.10.062 pmid:26518633 fatcat:grl77wcdwrdc7povgmypu4jibe

Type-II Errors of Independence Tests Can Lead to Arbitrarily Large Errors in Estimated Causal Effects: An Illustrative Example

Nicholas Cornia, Joris M. Mooij
2014 Conference on Uncertainty in Artificial Intelligence  
It is one of the simplest settings in which causal discovery and prediction methods based on conditional independences arrive at non-trivial conclusions, yet for which the lack of uniform consistency can  ...  Estimating the strength of causal effects from observational data is a common problem in scientific research.  ...  Acknowledgements We thank Tom Heskes for posing the problem, and Jonas Peters for inspiring discussions. We thank the reviewers for their comments that helped us improve the manuscript.  ... 
dblp:conf/uai/CorniaM14 fatcat:jmkyzbtvmrcjhjgvpcnanhnd4q

Computational causal discovery: Advantages and assumptions

Kun Zhang
2022 THEORIA : an International Journal for Theory, History and Fundations of Science  
The new contribution (Woodward, 2022) relies on that theory and further makes a big step towards empirical inference of causal relations from non-experimental data.  ...  In this paper, I will focus on some of the emerging computational methods for finding causal relations from non-experimental data and attempt to complement Woodward's contribution with discussions on 1  ...  R01HL159805, and by a grant from Apple. The United States Air Force or National Institutes of Health is not responsible for the views reported in this article.  ... 
doi:10.1387/theoria.22904 fatcat:ojivjn2rfrfvfas7l6igknek34

Perceiving the arrow of time in autoregressive motion

Kristof Meding, Dominik Janzing, Bernhard Schölkopf, Felix A. Wichmann
2019 Neural Information Processing Systems  
Understanding the principles of causal inference in the visual system has a long history at least since the seminal studies by Albert Michotte.  ...  Many cognitive and machine learning scientists believe that intelligent behavior requires agents to possess causal models of the world.  ...  in human vision (F.W. and B.S.).  ... 
dblp:conf/nips/MedingJSW19 fatcat:huqsxqybpjgi3mzsljh2s7shla
« Previous Showing results 1 — 15 out of 11,930 results