Learning the Structure of Linear Latent Variable Models

By Ricardo M. A. Silva, Richard Scheines, Clark Glymour, Peter Spirtes. 2006

286 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods, Statistical Learning Theory

How researchers see this paper

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

In Probabilistic & Bayesian Methods

In Statistical Learning Theory

Citations in the corpus

Causal Models for Dynamical Systems (2020, background): “…and on causal discovery [e.g. Spirtes et al., 2000, Pearl, 2009, Shpitser, 2018, Richardson and Spirtes, 2002, Hern´an and Robins, 2006, Evans, 2015, Richardson et al., 2017, Hyttinen et al., 2012, Verma and Pearl, 1991, Silva et al., 2006], more work is needed in the case of dynamical systems.”
Identifying Finite Mixtures of Nonparametric Product Distributions and Causal Inference of Confounders (2013, background): “Fast Causal Inference (Silva et al., 2006) can exclude confounding for some pairs of variables, given that many variables are observed.”
Comorbidity: A network perspective (2010, background): “The BPC (Build Pure Clusters) search algorithm (Silva et al. 2006) estimates whether a set of variables shares a latent common cause; BPC is asymptotically correct for binary variables whose values are two-valued projections of a Gaussian distribution.”
Gaussian Process Structural Equation Models with Latent Variables (2010, background): “Asparouhov and Muthén (2009) and Silva et al. (2006) discuss relevant issues.”
Discovering Hidden Variables in Noisy-Or Networks using Quartet Tests (2013, background): “Silva et al. (2006) study structure learning in linear models with continuous latent variables, giving an algorithm for discovering disjoint subsets of observed variables that have a single hidden variable as its parent.”
Nonlinear causal discovery with additive noise models (2008, background): “The development of causal discovery methods to infer causal relationships from uncontrolled data constitutes an important current research topic [1] [2] [3] [4] [5] [6] [7] [8].”
DirectLiNGAM: A Direct Method for Learning a Linear Non-Gaussian Structural Equation Model (2011, background): “5] and selected variances of the external inﬂuences e i from the interval [1 , 3] as in [27].”
Distinguishing Cause and Effect via Second Order Exponential Models (2009, methodology): “To show that F θ is non-singular we compute ∇ p θ ( y ) and ﬁnd points y 1 , . . . , y 4 and a value θ such that the corresponding gradients are linearly independent (see Appendix 7).”
A survey of Bayesian Network structure learning (2021, methodology): “Their method is based on the tetrad constraints (Silva et al., 2006) which can identify sets of four variables that are mutually d-separated by a common latent variable.”
Multiple Causes: A Causal Graphical View (2019, background): “A second body of related work is on causal structural learning with latent variables (Silva et al., 2006; Mckeigue et al., 2010; Anandkumar et al., 2014; Frot et al., 2017).”
Estimation of linear non-Gaussian acyclic models for latent factors (2009, methodology): “But many empirical researchers are more interested in discovering causal relations among latent factors [11].”
Estimation of causal effects using linear non-Gaussian causal models with hidden variables (2008, methodology): “The key difference to other existing research involving linear models with latent variables, such as [3,15], is our assumption of non-Gaussian variables, allowing us to utilize methods based on higher-order statistics.”
Temporal causal modeling with graphical granger methods (2007, background): “Indeed, this is the main idea behind many causal discovery algorithms such as the PC-algorithm [26, 1, 25].”
Causal Discovery with Linear Non-Gaussian Models under Measurement Error: Structural Identifiability Results (2018, methodology): “We note that in case multiple observed variables are generated from a single latent variable or a single observed variable is generated by multiple latent variables (see, e.g., Silva et al. (2006)), we can still use the CR-CAMME to represent the process.”
Search for Additive Nonlinear Time Series Causal Models (2008, background): “(Another group of causal inference algorithms that are based on model scores, such as Bayesian posteriors, are unable to handle either latent variables or feedbacks, except under extra constraints (Silva et al., 2006; Drton et al., 2006).”
Causal feature learning: an overview (2017, unknown): “Unobserved cognitive variables Li stand in some unknown causal relation (Reise et al. 2010; Silva et al. 2006).”