Results 11  20
of
57
Identifying confounders using additive noise models
, 2009
"... We propose a method for inferring the existence of a latent common cause (“confounder”) of two observed random variables. The method assumes that the two effects of the confounder are (possibly nonlinear) functions of the confounder plus independent, additive noise. We discuss under which conditions ..."
Abstract

Cited by 5 (5 self)
 Add to MetaCart
We propose a method for inferring the existence of a latent common cause (“confounder”) of two observed random variables. The method assumes that the two effects of the confounder are (possibly nonlinear) functions of the confounder plus independent, additive noise. We discuss under which conditions the model is identifiable (up to an arbitrary reparameterization of the confounder) from the joint distribution of the effects. We state and prove a theoretical result that provides evidence for the conjecture that the model is generically identifiable under suitable technical conditions. In addition, we propose a practical method to estimate the confounder from a finite i.i.d. sample of the effects and illustrate that the method works well on both simulated and realworld data.
Automated search for causal relations: Theory and practice
 Heuristics, Probability and Causality: A Tribute to Judea Pearl
, 2010
"... The rapid spread of interest in the last two decades in principled methods of search or estimation of causal relations has been driven in part by technological developments, especially the changing nature of modern data collection and storage techniques, and the increases in the speed and storage ca ..."
Abstract

Cited by 5 (0 self)
 Add to MetaCart
(Show Context)
The rapid spread of interest in the last two decades in principled methods of search or estimation of causal relations has been driven in part by technological developments, especially the changing nature of modern data collection and storage techniques, and the increases in the speed and storage capacities of computers. Statistics books from 30 years
What Is Right with ‘Bayes Net Methods’ and What Is Wrong with ‘Hunting Causes and Using Them’?, “The British Journal for the Philosophy of Science” 61.1
, 2010
"... NancyCartwright’s recent criticisms of efforts andmethods to obtain causal information from sample data using automated search are considered. In addition to reviewing that effort, I argue that almost all of her criticisms are false and rest on misreading, ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
NancyCartwright’s recent criticisms of efforts andmethods to obtain causal information from sample data using automated search are considered. In addition to reviewing that effort, I argue that almost all of her criticisms are false and rest on misreading,
Discriminative Mixtures of Sparse Latent Fields for Risk Management
"... We describe a simple and efficient approach to learning structures of sparse highdimensional latent variable models. Standard algorithms either learn structures of specific predefined forms, or estimate sparse graphs in the data space ignoring the possibility of the latent variables. In contrast, o ..."
Abstract

Cited by 4 (1 self)
 Add to MetaCart
(Show Context)
We describe a simple and efficient approach to learning structures of sparse highdimensional latent variable models. Standard algorithms either learn structures of specific predefined forms, or estimate sparse graphs in the data space ignoring the possibility of the latent variables. In contrast, our method learns rich dependencies and allows for latent variables that may confound the relations between the observations. We extend the model to conditional mixtures with side information and nonGaussian marginal distributions of the observations. We then show that our model may be used for learning sparse latent variable structures corresponding to multiple unknown states, and for uncovering features useful for explaining and predicting structural changes. We apply the model to realworld financial data with heavytailed marginals covering the low and high market volatility periods of 20052011. We show that our method tends to give rise to significantly higher likelihoods of test data than standard network learning methods exploiting the sparsity assumption. We also demonstrate that our approach may be practical for financial stresstesting and visualization of dependencies between financial instruments. 1
Node discovery problem for a social network, eprint arxiv.org/abs/0710
, 2007
"... This paper presents a practical heuristic algorithm to address a node discovery problem. The node discovery problem is to discover a clue on the person, who does not appear in the observed records, but is relevant functionally in affecting decisionmaking and behavior of an organization. We define t ..."
Abstract

Cited by 4 (3 self)
 Add to MetaCart
This paper presents a practical heuristic algorithm to address a node discovery problem. The node discovery problem is to discover a clue on the person, who does not appear in the observed records, but is relevant functionally in affecting decisionmaking and behavior of an organization. We define two topological relevance of a node in a social network (global and local relevance). Association between the topological relevance and the functional relevance is studied with a few example networks in criminal organizations. We propose a heuristic algorithm to infer an invisible, functionally relevant person. Its performance (precision, recall, and F value) is demonstrated with a simulation experiment using a network derived from the WattsStrogatz (WS) model. 1 Node discovery problem The activity of an organization is often under influence from an invisible relevant person. The term, invisible, means that the influence is not seen directly by the method applied in the observation procedure. This phenomenon arises intentionally or unintentionally. Let us show 2 examples. 1. Criminal organization: A commander tries to conceal himself from leaving any traces in communication and meeting logs, which are the basic intelligence to the police. Otherwise, exposure and arrest of a relevant pilot would have been a fatal damage to the terrorist organization in the 9/11 attack. 2. Manufacturing company: A sales person happens to be a close friend of an expertise factory engineer through a common friend: a product designer.
Joint estimation of linear nongaussian acyclic models
 Neurocomputing
, 2012
"... for latent factors ..."
(Show Context)
Gaussian process structural equation models with latent variables
 Proceedings of the 26th Conference on Uncertainty on Artificial Intelligence, UAI
, 2010
"... In a variety of disciplines such as social sciences, psychology, medicine and economics, the recorded data are considered to be noisy measurements of latent variables connected by some causal structure. This corresponds to a family of graphical models known as the structural equation model with late ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
In a variety of disciplines such as social sciences, psychology, medicine and economics, the recorded data are considered to be noisy measurements of latent variables connected by some causal structure. This corresponds to a family of graphical models known as the structural equation model with latent variables. While linear nonGaussian variants have been wellstudied, inference in nonparametric structural equation models is still underdeveloped. We introduce a sparse Gaussian process parameterization that defines a nonlinear structure connecting latent variables, unlike common formulations of Gaussian process latent variable models. The sparse parameterization is given a full Bayesian treatment without compromising Markov chain Monte Carlo efficiency. We compare the stability of the sampling procedure and the predictive ability of the model against the current practice. 1
Temporal Graphical Models for CrossSpecies Gene Regulatory Network Discovery
"... Many genes and biological processes function in similar ways across different species. Crossspecies gene expression analysis, as a powerful tool to characterize the dynamical properties of the cell, has found a number of applications, such as identifying a conserved core set of cell cycle genes. Ho ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
Many genes and biological processes function in similar ways across different species. Crossspecies gene expression analysis, as a powerful tool to characterize the dynamical properties of the cell, has found a number of applications, such as identifying a conserved core set of cell cycle genes. However, to the best of our knowledge, there is limited effort on developing appropriate techniques to capture the causal relations between genes from timeseries microarray data across species. In this paper, we present hidden Markov random field regression with L1 penalty to jointly uncover the regulatory networks for multiple species. The algorithm provides a framework for sharing information across species via hidden component graphs and can conveniently incorporate domain knowledge over evolution relationship between species. We demonstrate the effectiveness of our method on two synthetic datasets and one innate immune response microarray dataset. 1.
Constructing Variables that Support Causal Inference
, 2013
"... to many individuals. David Danks has been an ideal adviser and dissertation committee cochair. Despite a busy schedule and many advisees, he always found time to read drafts and provide careful feedback. His understanding ear and sage advice (both personal and professional) I value immensely. Thank ..."
Abstract

Cited by 2 (2 self)
 Add to MetaCart
to many individuals. David Danks has been an ideal adviser and dissertation committee cochair. Despite a busy schedule and many advisees, he always found time to read drafts and provide careful feedback. His understanding ear and sage advice (both personal and professional) I value immensely. Thank you. Richard Scheines, my other cochair, provided the intellectual inspiration for much of this work. His longstanding interests in applied causal inference and matters of variable construction and definition in the social sciences led him to ask challenging questions, seriously improving this work. My committee members have provided guidance and inspiration in a variety of ways. Daniel Neill, some time ago, stimulated my interest in developing interpretable models for predictive and causal inference that might prove useful for realworld policymakers. Partha Saha and Steven Ritter, in different settings and over the course of several years, have demonstrated that the type of questions we ask in this work are important in real educa
Latent Composite Likelihood Learning for the Structured Canonical Correlation Model: Supplementary Material
"... We present four pieces of supplementary material: first, an approach for the CDN inference problem of computing likelihood functions, which for our purposes we believe it is simpler to implement than other approaches presented in the literature; second, a discussion of the convergence of LEARNSTRUCT ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
We present four pieces of supplementary material: first, an approach for the CDN inference problem of computing likelihood functions, which for our purposes we believe it is simpler to implement than other approaches presented in the literature; second, a discussion of the convergence of LEARNSTRUCTUREDCCAII; third, brief comments on identification and initialization; fourth, details on the preprocessing of the NHS data. 1 SIMPLER CDN INFERENCE An efficient procedure for transforming CDFs into PMFs is given in detail by Huang et al. (2010), which is particularly sophisticated and seemingly hard to implement. However, one can reduce the problem of computing PMFs from CDFs following the structure of Equation (5) – itself just a rearrangement of the general formulation (Joe, 1997) for binary variables: just introduce “pseudo ” random variables corresponding to the difference indicatorsZand construct the corresponding factor graph. Notice that the term (−1) ∑p i=1 zi is itself a product of univariate factors over the pseudo set Z. Equation (5) is the “marginal ” of a pseudo distributionP(Z,Y) and can be found by any standard exact method of inference. We used junction trees. Figure 1 shows an example of reducing the problem of computing the PMF of graph Y1 ↔ Y2 ↔ Y3. The result is analogous in the continuous case: one just have to create indicator variables that pick which factors are being derived and which are not. This simple link is not mentioned in previous papers, to the best of our knowledge. In any case, the customized method described by Huang et al. (2010) readily includes details on how to generate parameter gradients, and it is useful as a framework for developing approximate algorithms (as already hinted by Huang and Frey, 2008): in our case, the