Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures

Size: px
Start display at page:

Download "Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures"

Transcription

1 Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures V. Asvatourian, S. Michiels, E. Lanoy Biostatistics and Epidemiology unit, Gustave-Roussy, Villejuif, France INSERM 1018 ONCOSTAT 5-6 October 2017 GDR Vahe Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 1

2 Current work Contribution of causal models in evaluating immunotherapies from observational data Motivations Multi-dimensional Predictive variables (p) > 400 Number of subjects (n) 40 Multi-collinearity Biomarkers measures can be repeated Observational data: association vs causation? Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 2

3 IDA: Estimation of causal effects in high-dimensionnal settings Maathuis, M. H., Kalisch, M., & Bühlmann, P. (2009). Estimating high-dimensional intervention effects from observational data. Annals of Statistics Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 3

4 Causal structure learning method Score-based To determine within candidate subgraphs those which fit better the data The fitness is measured through a score The purpose of the algorithm is to identify the DAG maximizing the score Time consuming in high-dimensional context Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 4

5 Causal structure learning method Score-based To determine within candidate subgraphs those which fit better the data The fitness is measured through a score The purpose of the algorithm is to identify the DAG maximizing the score Time consuming in high-dimensional context Constraint-based To learn the relationships by testing (conditionnal) independences between pairs of variables Consistent in high dimensional settings Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 4

6 Causal structure learning method PC-algorithm (constraint-based) Tests of (conditional) independences between pairs of variables Estimation of causal effects based on Completed partially DAG 1 X3 X11 X10 X4 X2 X12 X9 X5 X1 X13 X16 X6 X8 X14 X15 1 Maathuis, M. H., Kalisch, M., & Bühlmann, P. (2009). Estimating high-dimensional intervention effects from observational data. Annals of Statistics Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 5 X7

7 Causal structure learning method PC-algorithm (constraint-based) Tests of (conditional) independences between pairs of variables Estimation of causal effects based on Completed partially DAG 1 X3 X11 X10 X4 X2 X12 X9 X5 X1 X13 X16 X6 X8 X14 X15 1 Maathuis, M. H., Kalisch, M., & Bühlmann, P. (2009). Estimating high-dimensional intervention effects from observational data. Annals of Statistics Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 5 X7

8 Causal structure learning method PC-algorithm (constraint-based) Tests of (conditional) independences between pairs of variables Estimation of causal effects based on Completed partially DAG 1 X3 X11 X10 X4 X2 X12 X9 X5 X1 X13 X16 X6 X8 X14 X15 1 Maathuis, M. H., Kalisch, M., & Bühlmann, P. (2009). Estimating high-dimensional intervention effects from observational data. Annals of Statistics Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 5 X7

9 PC-algorithm on repeated exposures 1 Chronologically ordered PC-algorithm Integration of chronological order Estimation of causal effects based on Completed partially DAG X i,t=0 X i,t=1 X i,t=2 1 Asvatourian V, Coutzac C, Chaput N, Michiels S, Lanoy E: Estimating causal effects of repeated exposures on a binary endpoint in a high-dimensional setting Under review Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 6

10 PC-algorithm on repeated exposures 1 Chronologically ordered PC-algorithm Integration of chronological order Estimation of causal effects based on Completed partially DAG X i,t=0 X i,t=1 X i,t=2 1 Asvatourian V, Coutzac C, Chaput N, Michiels S, Lanoy E: Estimating causal effects of repeated exposures on a binary endpoint in a high-dimensional setting Under review Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 6

11 PC-algorithm on repeated exposures 1 Chronologically ordered PC-algorithm Integration of chronological order Estimation of causal effects based on Completed partially DAG X i,t=0 X i,t=1 X i,t=2 1 Asvatourian V, Coutzac C, Chaput N, Michiels S, Lanoy E: Estimating causal effects of repeated exposures on a binary endpoint in a high-dimensional setting Under review Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 6

12 PC-algorithm on repeated exposures 1 Chronologically ordered PC-algorithm Integration of chronological order Estimation of causal effects based on Completed partially DAG X i,t=0 X i,t=1 X i,t=2 1 Asvatourian V, Coutzac C, Chaput N, Michiels S, Lanoy E: Estimating causal effects of repeated exposures on a binary endpoint in a high-dimensional setting Under review Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 6

13 PC-algorithm on repeated exposures 1 Chronologically ordered PC-algorithm Integration of chronological order Estimation of causal effects based on Completed partially DAG X i,t=0 X i,t=1 X i,t=2 1 Asvatourian V, Coutzac C, Chaput N, Michiels S, Lanoy E: Estimating causal effects of repeated exposures on a binary endpoint in a high-dimensional setting Under review Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 6

14 Can we remove or fix edges (a priori information)? Some biomarkers could be known a priori Extensions of the score-based method Adding the expert opinion in the score calculation Reliable and consistent expert s opinion with the true structure Small number of experts Low dimensional setting Extensions of the constraint-based method Few articles on adding expert s knowledge in the case of constraint-based methods. Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 7

15 PC-algorithm with a priori information on high dimensional time-varying exposures? X i,t=0 X i,t=1 X i,t=2 X i,t=0 X i,t=1 X i,t=2 PC-algo without a priori information PC-algo with a priori information Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 8

16 PC-algorithm with a priori information on high dimensional time-varying exposures? X i,t=0 X i,t=1 X i,t=2 X i,t=0 X i,t=1 X i,t=2 PC-algo without a priori information PC-algo with a priori information Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 8

17 PC-algorithm with a priori information on high dimensional time-varying exposures? X i,t=0 X i,t=1 X i,t=2 X i,t=0 X i,t=1 X i,t=2 PC-algo without a priori information PC-algo with a priori information Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 8

18 Dynamic versus non-dynamic a priori information Giving information about time-varying biomarkers versus time fixed biomarkers X i,t=0 X i,t=1 X i,t=2 X i,t=0 X i,t=1 X i,t=2 Dynamic a priori information Non dynamic a priori information Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 9

19 Assumptions Is the true DAG dynamic or not? X i,t=0 X i,t=1 X i,t=2 X i,t=0 X i,t=1 X i,t=2 Dynamic true DAG Non dynamic true DAG Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 10

20 When to add a priori information? Steps of the PC-algorithm Identification of the skeleton Identification of the v-structures Orienting the last edges Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 11

21 When to add a priori information? Steps of the PC-algorithm Adding information about dependencies (fix or delete edges before performing PC-algo) Identification of the skeleton Adding information about direction (orient edges) Identification of the v-structures Orienting the last edges Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 11

22 Expert s knowledge elicitation For each pair, a priori information can be summarized by P A,B = {P, P, P } Each expert gives a score from 0 to 10 to every pairs he knows, where the score represents the confidence of causal effect If several scores are given to a same pair, the median of these score is taken Each expert is represented by 4 parameters γ 1, γ 2, γ 3, γ 4 : γ 1 γ 2 γ 3 γ 4 Probability of detecting the existing edges with correct direction Probability of detecting the existing edges with reverse direction Probability of correctly detecting the absent edges Probability of not detecting the absent edges Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 12

23 Simulation set-up p {20, 80} n visits {4, 8} Expert s level {bad, medium, good} Bad γ 1 : +, γ 2 : + + +, γ 3 : +, γ 4 : Medium γ 1 : ++, γ 2 : ++, γ 3 : ++, γ 4 : ++ Good γ 1 : + + +, γ 2 : +, γ 3 : + + +, γ 4 : + Percentage of a priori information {10, 40} n {50, 1000} Number of experts= 10 Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 13

24 Evaluation of performance The capacity of the algorithm to recover true edges through the sensitivity The capacity of the algorithm to recover true absences of edges through the specificity Structure similarity trough the Structural Hamming Distance (SHD), which is a score that compares the following errors Wrong connection: an absent edge in the original graph is present in the learned graph Missed edge: a true edge is not in the learned graph Wrong orientation: an edge has a different orientation in the original and learned structure Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 14

25 Discussion More simulations are ongoing Dynamic assumptions be not adequate in some biological settings Ability to identify true edges (sensitivity) remains low Perspective: Combining constraint-based and Bayesian methods Vahé Asvatourian Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures 15

Predicting causal effects in large-scale systems from observational data

Predicting causal effects in large-scale systems from observational data nature methods Predicting causal effects in large-scale systems from observational data Marloes H Maathuis 1, Diego Colombo 1, Markus Kalisch 1 & Peter Bühlmann 1,2 Supplementary figures and text: Supplementary

More information

Using background knowledge for the estimation of total causal e ects

Using background knowledge for the estimation of total causal e ects Using background knowledge for the estimation of total causal e ects Interpreting and using CPDAGs with background knowledge Emilija Perkovi, ETH Zurich Joint work with Markus Kalisch and Marloes Maathuis

More information

Causal Structure Learning and Inference: A Selective Review

Causal Structure Learning and Inference: A Selective Review Vol. 11, No. 1, pp. 3-21, 2014 ICAQM 2014 Causal Structure Learning and Inference: A Selective Review Markus Kalisch * and Peter Bühlmann Seminar for Statistics, ETH Zürich, CH-8092 Zürich, Switzerland

More information

arxiv: v1 [stat.ml] 23 Oct 2016

arxiv: v1 [stat.ml] 23 Oct 2016 Formulas for counting the sizes of Markov Equivalence Classes Formulas for Counting the Sizes of Markov Equivalence Classes of Directed Acyclic Graphs arxiv:1610.07921v1 [stat.ml] 23 Oct 2016 Yangbo He

More information

Learning With Bayesian Networks. Markus Kalisch ETH Zürich

Learning With Bayesian Networks. Markus Kalisch ETH Zürich Learning With Bayesian Networks Markus Kalisch ETH Zürich Inference in BNs - Review P(Burglary JohnCalls=TRUE, MaryCalls=TRUE) Exact Inference: P(b j,m) = c Sum e Sum a P(b)P(e)P(a b,e)p(j a)p(m a) Deal

More information

Graphical models and causality: Directed acyclic graphs (DAGs) and conditional (in)dependence

Graphical models and causality: Directed acyclic graphs (DAGs) and conditional (in)dependence Graphical models and causality: Directed acyclic graphs (DAGs) and conditional (in)dependence General overview Introduction Directed acyclic graphs (DAGs) and conditional independence DAGs and causal effects

More information

Supplementary material to Structure Learning of Linear Gaussian Structural Equation Models with Weak Edges

Supplementary material to Structure Learning of Linear Gaussian Structural Equation Models with Weak Edges Supplementary material to Structure Learning of Linear Gaussian Structural Equation Models with Weak Edges 1 PRELIMINARIES Two vertices X i and X j are adjacent if there is an edge between them. A path

More information

Causality. Bernhard Schölkopf and Jonas Peters MPI for Intelligent Systems, Tübingen. MLSS, Tübingen 21st July 2015

Causality. Bernhard Schölkopf and Jonas Peters MPI for Intelligent Systems, Tübingen. MLSS, Tübingen 21st July 2015 Causality Bernhard Schölkopf and Jonas Peters MPI for Intelligent Systems, Tübingen MLSS, Tübingen 21st July 2015 Charig et al.: Comparison of treatment of renal calculi by open surgery, (...), British

More information

Learning in Bayesian Networks

Learning in Bayesian Networks Learning in Bayesian Networks Florian Markowetz Max-Planck-Institute for Molecular Genetics Computational Molecular Biology Berlin Berlin: 20.06.2002 1 Overview 1. Bayesian Networks Stochastic Networks

More information

ANALYTIC COMPARISON. Pearl and Rubin CAUSAL FRAMEWORKS

ANALYTIC COMPARISON. Pearl and Rubin CAUSAL FRAMEWORKS ANALYTIC COMPARISON of Pearl and Rubin CAUSAL FRAMEWORKS Content Page Part I. General Considerations Chapter 1. What is the question? 16 Introduction 16 1. Randomization 17 1.1 An Example of Randomization

More information

Randomized dose-escalation design for drug combination cancer trials with immunotherapy

Randomized dose-escalation design for drug combination cancer trials with immunotherapy Randomized dose-escalation design for drug combination cancer trials with immunotherapy Pavel Mozgunov, Thomas Jaki, Xavier Paoletti Medical and Pharmaceutical Statistics Research Unit, Department of Mathematics

More information

Interpreting and using CPDAGs with background knowledge

Interpreting and using CPDAGs with background knowledge Interpreting and using CPDAGs with background knowledge Emilija Perković Seminar for Statistics ETH Zurich, Switzerland perkovic@stat.math.ethz.ch Markus Kalisch Seminar for Statistics ETH Zurich, Switzerland

More information

Probabilistic Causal Models

Probabilistic Causal Models Probabilistic Causal Models A Short Introduction Robin J. Evans www.stat.washington.edu/ rje42 ACMS Seminar, University of Washington 24th February 2011 1/26 Acknowledgements This work is joint with Thomas

More information

Bayesian Networks Basic and simple graphs

Bayesian Networks Basic and simple graphs Bayesian Networks Basic and simple graphs Ullrika Sahlin, Centre of Environmental and Climate Research Lund University, Sweden Ullrika.Sahlin@cec.lu.se http://www.cec.lu.se/ullrika-sahlin Bayesian [Belief]

More information

Causal inference (with statistical uncertainty) based on invariance: exploiting the power of heterogeneous data

Causal inference (with statistical uncertainty) based on invariance: exploiting the power of heterogeneous data Causal inference (with statistical uncertainty) based on invariance: exploiting the power of heterogeneous data Peter Bühlmann joint work with Jonas Peters Nicolai Meinshausen ... and designing new perturbation

More information

Towards an extension of the PC algorithm to local context-specific independencies detection

Towards an extension of the PC algorithm to local context-specific independencies detection Towards an extension of the PC algorithm to local context-specific independencies detection Feb-09-2016 Outline Background: Bayesian Networks The PC algorithm Context-specific independence: from DAGs to

More information

Causal Effect Evaluation and Causal Network Learning

Causal Effect Evaluation and Causal Network Learning and Peking University, China June 25, 2014 and Outline 1 Yule-Simpson paradox Causal effects Surrogate and surrogate paradox 2 and Outline Yule-Simpson paradox Causal effects Surrogate and surrogate paradox

More information

Protein Complex Identification by Supervised Graph Clustering

Protein Complex Identification by Supervised Graph Clustering Protein Complex Identification by Supervised Graph Clustering Yanjun Qi 1, Fernanda Balem 2, Christos Faloutsos 1, Judith Klein- Seetharaman 1,2, Ziv Bar-Joseph 1 1 School of Computer Science, Carnegie

More information

Local Structure Discovery in Bayesian Networks

Local Structure Discovery in Bayesian Networks 1 / 45 HELSINGIN YLIOPISTO HELSINGFORS UNIVERSITET UNIVERSITY OF HELSINKI Local Structure Discovery in Bayesian Networks Teppo Niinimäki, Pekka Parviainen August 18, 2012 University of Helsinki Department

More information

Inferring Causal Phenotype Networks from Segregating Populat

Inferring Causal Phenotype Networks from Segregating Populat Inferring Causal Phenotype Networks from Segregating Populations Elias Chaibub Neto chaibub@stat.wisc.edu Statistics Department, University of Wisconsin - Madison July 15, 2008 Overview Introduction Description

More information

arxiv: v3 [stat.me] 10 Mar 2016

arxiv: v3 [stat.me] 10 Mar 2016 Submitted to the Annals of Statistics ESTIMATING THE EFFECT OF JOINT INTERVENTIONS FROM OBSERVATIONAL DATA IN SPARSE HIGH-DIMENSIONAL SETTINGS arxiv:1407.2451v3 [stat.me] 10 Mar 2016 By Preetam Nandy,,

More information

Bounding the Probability of Causation in Mediation Analysis

Bounding the Probability of Causation in Mediation Analysis arxiv:1411.2636v1 [math.st] 10 Nov 2014 Bounding the Probability of Causation in Mediation Analysis A. P. Dawid R. Murtas M. Musio February 16, 2018 Abstract Given empirical evidence for the dependence

More information

Solutions to Odd-Numbered End-of-Chapter Exercises: Chapter 10

Solutions to Odd-Numbered End-of-Chapter Exercises: Chapter 10 Introduction to Econometrics (3 rd Updated Edition) by James H. Stock and Mark W. Watson Solutions to Odd-Numbered End-of-Chapter Exercises: Chapter 10 (This version July 20, 2014) Stock/Watson - Introduction

More information

Reductionist View: A Priori Algorithm and Vector-Space Text Retrieval. Sargur Srihari University at Buffalo The State University of New York

Reductionist View: A Priori Algorithm and Vector-Space Text Retrieval. Sargur Srihari University at Buffalo The State University of New York Reductionist View: A Priori Algorithm and Vector-Space Text Retrieval Sargur Srihari University at Buffalo The State University of New York 1 A Priori Algorithm for Association Rule Learning Association

More information

arxiv: v4 [math.st] 19 Jun 2018

arxiv: v4 [math.st] 19 Jun 2018 Complete Graphical Characterization and Construction of Adjustment Sets in Markov Equivalence Classes of Ancestral Graphs Emilija Perković, Johannes Textor, Markus Kalisch and Marloes H. Maathuis arxiv:1606.06903v4

More information

arxiv: v1 [cs.lg] 26 May 2017

arxiv: v1 [cs.lg] 26 May 2017 Learning Causal tructures Using Regression Invariance ariv:1705.09644v1 [cs.lg] 26 May 2017 AmirEmad Ghassami, aber alehkaleybar, Negar Kiyavash, Kun Zhang Department of ECE, University of Illinois at

More information

This paper has been submitted for consideration for publication in Biometrics

This paper has been submitted for consideration for publication in Biometrics Biometrics 63, 1?? March 2014 DOI: 10.1111/j.1541-0420.2005.00454.x PenPC: A Two-step Approach to Estimate the Skeletons of High Dimensional Directed Acyclic Graphs Min Jin Ha Department of Biostatistics,

More information

Logic, Knowledge Representation and Bayesian Decision Theory

Logic, Knowledge Representation and Bayesian Decision Theory Logic, Knowledge Representation and Bayesian Decision Theory David Poole University of British Columbia Overview Knowledge representation, logic, decision theory. Belief networks Independent Choice Logic

More information

Causal Effect Identification in Alternative Acyclic Directed Mixed Graphs

Causal Effect Identification in Alternative Acyclic Directed Mixed Graphs Proceedings of Machine Learning Research vol 73:21-32, 2017 AMBN 2017 Causal Effect Identification in Alternative Acyclic Directed Mixed Graphs Jose M. Peña Linköping University Linköping (Sweden) jose.m.pena@liu.se

More information

Inferring Protein-Signaling Networks

Inferring Protein-Signaling Networks Inferring Protein-Signaling Networks Lectures 14 Nov 14, 2011 CSE 527 Computational Biology, Fall 2011 Instructor: Su-In Lee TA: Christopher Miles Monday & Wednesday 12:00-1:20 Johnson Hall (JHN) 022 1

More information

Markovian Combination of Decomposable Model Structures: MCMoSt

Markovian Combination of Decomposable Model Structures: MCMoSt Markovian Combination 1/45 London Math Society Durham Symposium on Mathematical Aspects of Graphical Models. June 30 - July, 2008 Markovian Combination of Decomposable Model Structures: MCMoSt Sung-Ho

More information

Robust Monte Carlo Methods for Sequential Planning and Decision Making

Robust Monte Carlo Methods for Sequential Planning and Decision Making Robust Monte Carlo Methods for Sequential Planning and Decision Making Sue Zheng, Jason Pacheco, & John Fisher Sensing, Learning, & Inference Group Computer Science & Artificial Intelligence Laboratory

More information

COMP538: Introduction to Bayesian Networks

COMP538: Introduction to Bayesian Networks COMP538: Introduction to Bayesian Networks Lecture 9: Optimal Structure Learning Nevin L. Zhang lzhang@cse.ust.hk Department of Computer Science and Engineering Hong Kong University of Science and Technology

More information

Beyond Uniform Priors in Bayesian Network Structure Learning

Beyond Uniform Priors in Bayesian Network Structure Learning Beyond Uniform Priors in Bayesian Network Structure Learning (for Discrete Bayesian Networks) scutari@stats.ox.ac.uk Department of Statistics April 5, 2017 Bayesian Network Structure Learning Learning

More information

Computational Genomics. Systems biology. Putting it together: Data integration using graphical models

Computational Genomics. Systems biology. Putting it together: Data integration using graphical models 02-710 Computational Genomics Systems biology Putting it together: Data integration using graphical models High throughput data So far in this class we discussed several different types of high throughput

More information

Bayesian networks as causal models. Peter Antal

Bayesian networks as causal models. Peter Antal Bayesian networks as causal models Peter Antal antal@mit.bme.hu A.I. 3/20/2018 1 Can we represent exactly (in)dependencies by a BN? From a causal model? Suff.&nec.? Can we interpret edges as causal relations

More information

Identification and Estimation of Causal Effects from Dependent Data

Identification and Estimation of Causal Effects from Dependent Data Identification and Estimation of Causal Effects from Dependent Data Eli Sherman esherman@jhu.edu with Ilya Shpitser Johns Hopkins Computer Science 12/6/2018 Eli Sherman Identification and Estimation of

More information

Lecture 5: Bayesian Network

Lecture 5: Bayesian Network Lecture 5: Bayesian Network Topics of this lecture What is a Bayesian network? A simple example Formal definition of BN A slightly difficult example Learning of BN An example of learning Important topics

More information

Causal Inference. Prediction and causation are very different. Typical questions are:

Causal Inference. Prediction and causation are very different. Typical questions are: Causal Inference Prediction and causation are very different. Typical questions are: Prediction: Predict Y after observing X = x Causation: Predict Y after setting X = x. Causation involves predicting

More information

Challenges (& Some Solutions) and Making Connections

Challenges (& Some Solutions) and Making Connections Challenges (& Some Solutions) and Making Connections Real-life Search All search algorithm theorems have form: If the world behaves like, then (probability 1) the algorithm will recover the true structure

More information

Causal Bayesian networks. Peter Antal

Causal Bayesian networks. Peter Antal Causal Bayesian networks Peter Antal antal@mit.bme.hu A.I. 4/8/2015 1 Can we represent exactly (in)dependencies by a BN? From a causal model? Suff.&nec.? Can we interpret edges as causal relations with

More information

An Empirical-Bayes Score for Discrete Bayesian Networks

An Empirical-Bayes Score for Discrete Bayesian Networks An Empirical-Bayes Score for Discrete Bayesian Networks scutari@stats.ox.ac.uk Department of Statistics September 8, 2016 Bayesian Network Structure Learning Learning a BN B = (G, Θ) from a data set D

More information

CS 188: Artificial Intelligence. Outline

CS 188: Artificial Intelligence. Outline CS 188: Artificial Intelligence Lecture 21: Perceptrons Pieter Abbeel UC Berkeley Many slides adapted from Dan Klein. Outline Generative vs. Discriminative Binary Linear Classifiers Perceptron Multi-class

More information

Abstract. Three Methods and Their Limitations. N-1 Experiments Suffice to Determine the Causal Relations Among N Variables

Abstract. Three Methods and Their Limitations. N-1 Experiments Suffice to Determine the Causal Relations Among N Variables N-1 Experiments Suffice to Determine the Causal Relations Among N Variables Frederick Eberhardt Clark Glymour 1 Richard Scheines Carnegie Mellon University Abstract By combining experimental interventions

More information

Extreme values: Maxima and minima. October 16, / 12

Extreme values: Maxima and minima. October 16, / 12 Extreme values: Maxima and minima October 16, 2015 1 / 12 Motivation for Maxima and Minima Imagine you have some model which predicts profits / cost of doing something /.... Then you probably want to find

More information

Copula PC Algorithm for Causal Discovery from Mixed Data

Copula PC Algorithm for Causal Discovery from Mixed Data Copula PC Algorithm for Causal Discovery from Mixed Data Ruifei Cui ( ), Perry Groot, and Tom Heskes Institute for Computing and Information Sciences, Radboud University, Nijmegen, The Netherlands {R.Cui,

More information

Statistical Models for Causal Analysis

Statistical Models for Causal Analysis Statistical Models for Causal Analysis Teppei Yamamoto Keio University Introduction to Causal Inference Spring 2016 Three Modes of Statistical Inference 1. Descriptive Inference: summarizing and exploring

More information

A time series causal model

A time series causal model MPRA Munich Personal RePEc Archive A time series causal model Pu Chen Melbourne University September 2010 Online at http://mpra.ub.uni-muenchen.de/24841/ MPRA Paper No. 24841, posted 13. September 2010

More information

DEALING WITH MULTIVARIATE OUTCOMES IN STUDIES FOR CAUSAL EFFECTS

DEALING WITH MULTIVARIATE OUTCOMES IN STUDIES FOR CAUSAL EFFECTS DEALING WITH MULTIVARIATE OUTCOMES IN STUDIES FOR CAUSAL EFFECTS Donald B. Rubin Harvard University 1 Oxford Street, 7th Floor Cambridge, MA 02138 USA Tel: 617-495-5496; Fax: 617-496-8057 email: rubin@stat.harvard.edu

More information

A Brief Introduction to Graphical Models. Presenter: Yijuan Lu November 12,2004

A Brief Introduction to Graphical Models. Presenter: Yijuan Lu November 12,2004 A Brief Introduction to Graphical Models Presenter: Yijuan Lu November 12,2004 References Introduction to Graphical Models, Kevin Murphy, Technical Report, May 2001 Learning in Graphical Models, Michael

More information

Alexina Mason. Department of Epidemiology and Biostatistics Imperial College, London. 16 February 2010

Alexina Mason. Department of Epidemiology and Biostatistics Imperial College, London. 16 February 2010 Strategy for modelling non-random missing data mechanisms in longitudinal studies using Bayesian methods: application to income data from the Millennium Cohort Study Alexina Mason Department of Epidemiology

More information

Data Mining 2018 Bayesian Networks (1)

Data Mining 2018 Bayesian Networks (1) Data Mining 2018 Bayesian Networks (1) Ad Feelders Universiteit Utrecht Ad Feelders ( Universiteit Utrecht ) Data Mining 1 / 49 Do you like noodles? Do you like noodles? Race Gender Yes No Black Male 10

More information

Controlling for latent confounding by confirmatory factor analysis (CFA) Blinded Blinded

Controlling for latent confounding by confirmatory factor analysis (CFA) Blinded Blinded Controlling for latent confounding by confirmatory factor analysis (CFA) Blinded Blinded 1 Background Latent confounder is common in social and behavioral science in which most of cases the selection mechanism

More information

arxiv: v6 [math.st] 3 Feb 2018

arxiv: v6 [math.st] 3 Feb 2018 Submitted to the Annals of Statistics HIGH-DIMENSIONAL CONSISTENCY IN SCORE-BASED AND HYBRID STRUCTURE LEARNING arxiv:1507.02608v6 [math.st] 3 Feb 2018 By Preetam Nandy,, Alain Hauser and Marloes H. Maathuis,

More information

Lecture 6: Graphical Models: Learning

Lecture 6: Graphical Models: Learning Lecture 6: Graphical Models: Learning 4F13: Machine Learning Zoubin Ghahramani and Carl Edward Rasmussen Department of Engineering, University of Cambridge February 3rd, 2010 Ghahramani & Rasmussen (CUED)

More information

CAUSAL GAN: LEARNING CAUSAL IMPLICIT GENERATIVE MODELS WITH ADVERSARIAL TRAINING

CAUSAL GAN: LEARNING CAUSAL IMPLICIT GENERATIVE MODELS WITH ADVERSARIAL TRAINING CAUSAL GAN: LEARNING CAUSAL IMPLICIT GENERATIVE MODELS WITH ADVERSARIAL TRAINING (Murat Kocaoglu, Christopher Snyder, Alexandros G. Dimakis & Sriram Vishwanath, 2017) Summer Term 2018 Created for the Seminar

More information

arxiv: v1 [cs.ai] 16 Aug 2016

arxiv: v1 [cs.ai] 16 Aug 2016 Evaluating Causal Models by Comparing Interventional Distributions arxiv:1608.04698v1 [cs.ai] 16 Aug 2016 Dan Garant dgarant@cs.umass.edu College of Information and Computer Sciences University of Massachusetts

More information

Optimal Treatment Regimes for Survival Endpoints from a Classification Perspective. Anastasios (Butch) Tsiatis and Xiaofei Bai

Optimal Treatment Regimes for Survival Endpoints from a Classification Perspective. Anastasios (Butch) Tsiatis and Xiaofei Bai Optimal Treatment Regimes for Survival Endpoints from a Classification Perspective Anastasios (Butch) Tsiatis and Xiaofei Bai Department of Statistics North Carolina State University 1/35 Optimal Treatment

More information

Machine Learning Summer School

Machine Learning Summer School Machine Learning Summer School Lecture 3: Learning parameters and structure Zoubin Ghahramani zoubin@eng.cam.ac.uk http://learning.eng.cam.ac.uk/zoubin/ Department of Engineering University of Cambridge,

More information

ESTIMATING HIGH-DIMENSIONAL INTERVENTION EFFECTS FROM OBSERVATIONAL DATA

ESTIMATING HIGH-DIMENSIONAL INTERVENTION EFFECTS FROM OBSERVATIONAL DATA The Annals of Statistics 2009, Vol. 37, No. 6A, 3133 3164 DOI: 10.1214/09-AOS685 Institute of Mathematical Statistics, 2009 ESTIMATING HIGH-DIMENSIONAL INTERVENTION EFFECTS FROM OBSERVATIONAL DATA BY MARLOES

More information

Robustification of the PC-algorithm for Directed Acyclic Graphs

Robustification of the PC-algorithm for Directed Acyclic Graphs Robustification of the PC-algorithm for Directed Acyclic Graphs Markus Kalisch, Peter Bühlmann May 26, 2008 Abstract The PC-algorithm ([13]) was shown to be a powerful method for estimating the equivalence

More information

From Causality, Second edition, Contents

From Causality, Second edition, Contents From Causality, Second edition, 2009. Preface to the First Edition Preface to the Second Edition page xv xix 1 Introduction to Probabilities, Graphs, and Causal Models 1 1.1 Introduction to Probability

More information

An Introduction to Reversible Jump MCMC for Bayesian Networks, with Application

An Introduction to Reversible Jump MCMC for Bayesian Networks, with Application An Introduction to Reversible Jump MCMC for Bayesian Networks, with Application, CleverSet, Inc. STARMAP/DAMARS Conference Page 1 The research described in this presentation has been funded by the U.S.

More information

AP CALCULUS AB 2006 SCORING GUIDELINES (Form B) Question 2. the

AP CALCULUS AB 2006 SCORING GUIDELINES (Form B) Question 2. the AP CALCULUS AB 2006 SCORING GUIDELINES (Form B) Question 2 Let f be the function defined for x 0 with f ( 0) = 5 and f, the ( x 4) 2 first derivative of f, given by f ( x) = e sin ( x ). The graph of y

More information

Bayesian Networks to design optimal experiments. Davide De March

Bayesian Networks to design optimal experiments. Davide De March Bayesian Networks to design optimal experiments Davide De March davidedemarch@gmail.com 1 Outline evolutionary experimental design in high-dimensional space and costly experimentation the microwell mixture

More information

Empirical Bayes Moderation of Asymptotically Linear Parameters

Empirical Bayes Moderation of Asymptotically Linear Parameters Empirical Bayes Moderation of Asymptotically Linear Parameters Nima Hejazi Division of Biostatistics University of California, Berkeley stat.berkeley.edu/~nhejazi nimahejazi.org twitter/@nshejazi github/nhejazi

More information

Post-exam 2 practice questions 18.05, Spring 2014

Post-exam 2 practice questions 18.05, Spring 2014 Post-exam 2 practice questions 18.05, Spring 2014 Note: This is a set of practice problems for the material that came after exam 2. In preparing for the final you should use the previous review materials,

More information

Foundation Unit 16 topic test

Foundation Unit 16 topic test Name: Foundation Unit 16 topic test Date: Time: 40 minutes Total marks available: 34 Total marks achieved: Questions Q1. (a) Simplify c + c + c + c (1) (b) Simplify 6 m 5 (1) (c) Simplify 2e 3f + 7e 5f

More information

Causal Bayesian networks. Peter Antal

Causal Bayesian networks. Peter Antal Causal Bayesian networks Peter Antal antal@mit.bme.hu A.I. 11/25/2015 1 Can we represent exactly (in)dependencies by a BN? From a causal model? Suff.&nec.? Can we interpret edges as causal relations with

More information

Discovering Causal Structure from Observations

Discovering Causal Structure from Observations Chapter 24 Discovering Causal Structure from Observations The last few chapters have, hopefully, convinced you that when you want to do causal inference, knowing the causal graph is very helpful. We have

More information

Introduction to Causal Calculus

Introduction to Causal Calculus Introduction to Causal Calculus Sanna Tyrväinen University of British Columbia August 1, 2017 1 / 1 2 / 1 Bayesian network Bayesian networks are Directed Acyclic Graphs (DAGs) whose nodes represent random

More information

A review of some recent advances in causal inference

A review of some recent advances in causal inference A review of some recent advances in causal inference Marloes H. Maathuis and Preetam Nandy arxiv:1506.07669v1 [stat.me] 25 Jun 2015 Contents 1 Introduction 1 1.1 Causal versus non-causal research questions.................

More information

Markov properties for directed graphs

Markov properties for directed graphs Graphical Models, Lecture 7, Michaelmas Term 2009 November 2, 2009 Definitions Structural relations among Markov properties Factorization G = (V, E) simple undirected graph; σ Say σ satisfies (P) the pairwise

More information

Embellishing a Bayesian Network using a Chain Event Graph

Embellishing a Bayesian Network using a Chain Event Graph Embellishing a Bayesian Network using a Chain Event Graph L. M. Barclay J. L. Hutton J. Q. Smith University of Warwick 4th Annual Conference of the Australasian Bayesian Network Modelling Society Example

More information

arxiv: v1 [stat.ml] 16 Jun 2018

arxiv: v1 [stat.ml] 16 Jun 2018 The Reduced PC-Algorithm: Improved Causal Structure Learning in Large Random Networks Arjun Sondhi Department of Biostatistics University of Washington asondhi@uw.edu Ali Shojaie Department of Biostatistics

More information

Fast Adaptive Algorithm for Robust Evaluation of Quality of Experience

Fast Adaptive Algorithm for Robust Evaluation of Quality of Experience Fast Adaptive Algorithm for Robust Evaluation of Quality of Experience Qianqian Xu, Ming Yan, Yuan Yao October 2014 1 Motivation Mean Opinion Score vs. Paired Comparisons Crowdsourcing Ranking on Internet

More information

arxiv: v1 [cs.lg] 11 Jul 2018

arxiv: v1 [cs.lg] 11 Jul 2018 Causal discovery in the presence of missing data arxiv:1807.04010v1 [cs.lg] 11 Jul 2018 Ruibo Tu Cheng Zhang Paul Ackermann Hedvig Kjellström Kun Zhang Abstract Missing data are ubiquitous in many domains

More information

Data-driven discovery of modulatory factors for African rainfall variability

Data-driven discovery of modulatory factors for African rainfall variability Data-driven discovery of modulatory factors for African rainfall variability Michael Angus Gonzalo Bello Mandar Chaudhary North Carolina State University Fifth Workshop on Understanding Climate Change

More information

Targeted Maximum Likelihood Estimation in Safety Analysis

Targeted Maximum Likelihood Estimation in Safety Analysis Targeted Maximum Likelihood Estimation in Safety Analysis Sam Lendle 1 Bruce Fireman 2 Mark van der Laan 1 1 UC Berkeley 2 Kaiser Permanente ISPE Advanced Topics Session, Barcelona, August 2012 1 / 35

More information

Structure learning in human causal induction

Structure learning in human causal induction Structure learning in human causal induction Joshua B. Tenenbaum & Thomas L. Griffiths Department of Psychology Stanford University, Stanford, CA 94305 jbt,gruffydd @psych.stanford.edu Abstract We use

More information

arxiv: v3 [stat.me] 3 Jun 2015

arxiv: v3 [stat.me] 3 Jun 2015 The Annals of Statistics 2015, Vol. 43, No. 3, 1060 1088 DOI: 10.1214/14-AOS1295 c Institute of Mathematical Statistics, 2015 A GENERALIZED BACK-DOOR CRITERION 1 arxiv:1307.5636v3 [stat.me] 3 Jun 2015

More information

Comments on The Role of Large Scale Assessments in Research on Educational Effectiveness and School Development by Eckhard Klieme, Ph.D.

Comments on The Role of Large Scale Assessments in Research on Educational Effectiveness and School Development by Eckhard Klieme, Ph.D. Comments on The Role of Large Scale Assessments in Research on Educational Effectiveness and School Development by Eckhard Klieme, Ph.D. David Kaplan Department of Educational Psychology The General Theme

More information

Persuading Skeptics and Reaffirming Believers

Persuading Skeptics and Reaffirming Believers Persuading Skeptics and Reaffirming Believers May, 31 st, 2014 Becker-Friedman Institute Ricardo Alonso and Odilon Camara Marshall School of Business - USC Introduction Sender wants to influence decisions

More information

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2014

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2014 Bayesian Networks: Construction, Inference, Learning and Causal Interpretation Volker Tresp Summer 2014 1 Introduction So far we were mostly concerned with supervised learning: we predicted one or several

More information

What Causality Is (stats for mathematicians)

What Causality Is (stats for mathematicians) What Causality Is (stats for mathematicians) Andrew Critch UC Berkeley August 31, 2011 Introduction Foreword: The value of examples With any hard question, it helps to start with simple, concrete versions

More information

Bayesian Networks. Motivation

Bayesian Networks. Motivation Bayesian Networks Computer Sciences 760 Spring 2014 http://pages.cs.wisc.edu/~dpage/cs760/ Motivation Assume we have five Boolean variables,,,, The joint probability is,,,, How many state configurations

More information

Support Vector Machines (SVM) in bioinformatics. Day 1: Introduction to SVM

Support Vector Machines (SVM) in bioinformatics. Day 1: Introduction to SVM 1 Support Vector Machines (SVM) in bioinformatics Day 1: Introduction to SVM Jean-Philippe Vert Bioinformatics Center, Kyoto University, Japan Jean-Philippe.Vert@mines.org Human Genome Center, University

More information

Teaching a Prestatistics Course: Propelling Non-STEM Students Forward

Teaching a Prestatistics Course: Propelling Non-STEM Students Forward Teaching a Prestatistics Course: Propelling Non-STEM Students Forward Jay Lehmann College of San Mateo MathNerdJay@aol.com www.pearsonhighered.com/lehmannseries Learning Is in the Details Detailing concepts

More information

Causal Inference for High-Dimensional Data. Atlantic Causal Conference

Causal Inference for High-Dimensional Data. Atlantic Causal Conference Causal Inference for High-Dimensional Data Atlantic Causal Conference Overview Conditional Independence Directed Acyclic Graph (DAG) Models factorization and d-separation Markov equivalence Structure Learning

More information

arxiv: v1 [math.st] 13 Mar 2013

arxiv: v1 [math.st] 13 Mar 2013 Jointly interventional and observational data: estimation of interventional Markov equivalence classes of directed acyclic graphs arxiv:1303.316v1 [math.st] 13 Mar 013 Alain Hauser and Peter Bühlmann {hauser,

More information

Lecture 24, Causal Discovery

Lecture 24, Causal Discovery Lecture 24, Causal Discovery 36-402, Advanced Data Analysis 21 April 2011 Contents 1 Testing DAGs 1 2 Causal Discovery with Known Variables 3 2.1 Causal Discovery with Hidden Variables.............. 5

More information

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

Type-II Errors of Independence Tests Can Lead to Arbitrarily Large Errors in Estimated Causal Effects: An Illustrative Example Type-II Errors of Independence Tests Can Lead to Arbitrarily Large Errors in Estimated Causal Effects: An Illustrative Example Nicholas Cornia & Joris M. Mooij Informatics Institute University of Amsterdam,

More information

Environmentally-induced Population Displacements: Conclusions from PERN s Online Seminar

Environmentally-induced Population Displacements: Conclusions from PERN s Online Seminar Environment, Forced Migration & Social Vulnerability International Conference 9-11 October 2008 Bonn, Germany www.efmsv2008.org Environmentally-induced Population Displacements: Conclusions from PERN s

More information

How Mega is the Mega? Measuring the Spillover Effects of WeChat by Machine Learning and Econometrics

How Mega is the Mega? Measuring the Spillover Effects of WeChat by Machine Learning and Econometrics How Mega is the Mega? Measuring the Spillover Effects of WeChat by Machine Learning and Econometrics Jinyang Zheng Michael G. Foster School of Business, University of Washington, zhengjy@uw.edu Zhengling

More information

Adaptive Trial Designs

Adaptive Trial Designs Adaptive Trial Designs Wenjing Zheng, Ph.D. Methods Core Seminar Center for AIDS Prevention Studies University of California, San Francisco Nov. 17 th, 2015 Trial Design! Ethical:!eg.! Safety!! Efficacy!

More information

Impact Evaluation of Mindspark Centres

Impact Evaluation of Mindspark Centres Impact Evaluation of Mindspark Centres March 27th, 2014 Executive Summary About Educational Initiatives and Mindspark Educational Initiatives (EI) is a prominent education organization in India with the

More information

GLOBEX Bioinformatics (Summer 2015) Genetic networks and gene expression data

GLOBEX Bioinformatics (Summer 2015) Genetic networks and gene expression data GLOBEX Bioinformatics (Summer 2015) Genetic networks and gene expression data 1 Gene Networks Definition: A gene network is a set of molecular components, such as genes and proteins, and interactions between

More information

An extrapolation framework to specify requirements for drug development in children

An extrapolation framework to specify requirements for drug development in children An framework to specify requirements for drug development in children Martin Posch joint work with Gerald Hlavin Franz König Christoph Male Peter Bauer Medical University of Vienna Clinical Trials in Small

More information

Constraint-based Causal Discovery with Mixed Data

Constraint-based Causal Discovery with Mixed Data Noname manuscript No. (will be inserted by the editor) Constraint-based Causal Discovery with Mixed Data Michail Tsagris Giorgos Borboudakis Vincenzo Lagani Ioannis Tsamardinos Received: date / Accepted:

More information

AP CALCULUS AB 2011 SCORING GUIDELINES

AP CALCULUS AB 2011 SCORING GUIDELINES AP CALCULUS AB 2011 SCORING GUIDELINES Question 4 The continuous function f is defined on the interval 4 x 3. The graph of f consists of two quarter circles and one line segment, as shown in the figure

More information