Bayesian Inference for Contact Networks Given Epidemic Data

Size: px
Start display at page:

Download "Bayesian Inference for Contact Networks Given Epidemic Data"

Transcription

1 Bayesian Inference for Contact Networks Given Epidemic Data Chris Groendyke, David Welch, Shweta Bansal, David Hunter Departments of Statistics and Biology Pennsylvania State University SAMSI, April 17, 010 Supported by NIH Grant R01-GM

2 Outline 1 Inference for Contact Networks Epidemic Data 3 Simulation studies 4 Hagelloch Measles Data 5 Future Extensions

3 Outline 1 Inference for Contact Networks Epidemic Data 3 Simulation studies 4 Hagelloch Measles Data 5 Future Extensions

4 General goal Contact Network Nodes represent individuals; edges represent potentially disease-causing contacts between two individuals (context-dependent). Given an epidemic in a population transmitted across a (generally unobserved) contact network, we d like to be able to describe the properties of this network. NB: Obtaining the contact network itself is not necessarily a goal.

5 Contact networks and transmission networks Assume a contact network G on the individuals: A contact is necessary for disease transmission

6 Contact networks and transmission networks E Assume a contact network G on the individuals: A contact is necessary for disease transmission. Beginning with the first infected, disease is spread at exponential rate β, defining a subtree of the contact network called the transmission tree P. Data E 7, E 6,... are exposure times.

7 Contact networks and transmission networks 8 4 E E Assume a contact network G on the individuals: A contact is necessary for disease transmission. Beginning with the first infected, disease is spread at exponential rate β, defining a subtree of the contact network called the transmission tree P. Data E 7, E 6,... are exposure times.

8 Contact networks and transmission networks 8 4 E 5 E E Assume a contact network G on the individuals: A contact is necessary for disease transmission. Beginning with the first infected, disease is spread at exponential rate β, defining a subtree of the contact network called the transmission tree P. Data E 7, E 6,... are exposure times.

9 Contact networks and transmission networks 8 4 E 5 E E E 3 Assume a contact network G on the individuals: A contact is necessary for disease transmission. Beginning with the first infected, disease is spread at exponential rate β, defining a subtree of the contact network called the transmission tree P. Data E 7, E 6,... are exposure times.

10 Contact networks and transmission networks 8 4 E E 5 E E E 3 Assume a contact network G on the individuals: A contact is necessary for disease transmission. Beginning with the first infected, disease is spread at exponential rate β, defining a subtree of the contact network called the transmission tree P. Data E 7, E 6,... are exposure times.

11 Contact networks and transmission networks 8 4 E E 5 E E E 3 Assume a contact network G on the individuals: A contact is necessary for disease transmission. Beginning with the first infected, disease is spread at exponential rate β, defining a subtree of the contact network called the transmission tree P. Data E 7, E 6,... are exposure times. Simplistic representation of the transmission tree, P:

12 Existing literature We will use data from an epidemic to perform simultaneous inference on the network and epidemic parameters. A few papers (Britton and O Neill (00), Neal and Roberts (005), Ray and Marzouk (008)) have discussed this type of inference. These papers make very significant simplifying assumptions, and no papers have attempted to use more general network models or analyze larger data sets. Here, we will use Britton and O Neill (00) as a starting point...

13 Statistical vs. Probabilistic Modeling paradigm: Probability: Simulate networks from model, epidemic data on network Statistics (probability in reverse): Start with epidemic data, learn about parameters via understanding of model!

14 ERGMs The framework we use to model contact networks is the Exponential-family Random Graph Model (ERGM) or where P η (Y = y) exp{η t g(y)} P η (Y = y) = exp{ηt g(y)}, κ(η) η is a vector of parameters g(y) is a known vector of graph statistics on y κ(η) is the normalizing constant

15 Erdős - Rényi network model Let g(g) = G consist of the single statistic counting the number of edges in G. This gives as an ERGM where the (scalar) parameter P η (Y = y) = exp{ηg}, κ(η) η = logit(p) = log ( p ) 1 p is the log-odds of the existence of an edge. NB: We ll use p, not η, throughout.

16 A few words about software R: An open-source statistical package statnet: An R package for network analysis See volume 4 of Journal of Statistical Software Methods described later are in R package epinet.

17 Outline 1 Inference for Contact Networks Epidemic Data 3 Simulation studies 4 Hagelloch Measles Data 5 Future Extensions

18 Compartmental Models This type of model partitions the population into multiple classes, based on current disease status. One type of compartmental model is the SIR model: Susceptible Infective Removed The SEIR model adds an Exposed class, corresponding to a latent period for the disease: Susceptible Exposed Infective Removed

19 Compartmental Models This type of model partitions the population into multiple classes, based on current disease status. One type of compartmental model is the SIR model: Susceptible Infective Removed The SEIR model adds an Exposed class, corresponding to a latent period for the disease: Susceptible Exposed Infective Removed

20 Example Toy Dataset (ideal) Node Exposure Time Infective Time Removal Time Contact Network Transmission Tree

21 The loglikelihood: EIR times observed Parameters: β, k, θ, η Data: E, I, R, (G, P) L(parameters) = f (E, I, R, G, P β, k, θ, η) G,P = G,P f (E, I, R β, k, θ, G, P)f (P G)f (G η), where:

22 The loglikelihood: EIR times observed Parameters: β, k, θ, η Data: E, I, R, (G, P) L(parameters) = f (E, I, R, G, P β, k, θ, η) G,P = G,P f (E, I, R β, k, θ, G, P)f (P G)f (G η), f (E, I, R β, k, θ, G, P) models for how the times (Exposed, Infected, Recovered) depend on the transmission parameters and the networks G and P: For each infected i, E i is determined by β and G and P, while I i E i Gamma(k E, θ E ) and R i I i Gamma(k I, θ I ).

23 The loglikelihood: EIR times observed Parameters: β, k, θ, η Data: E, I, R, (G, P) L(parameters) = f (E, I, R, G, P β, k, θ, η) G,P = G,P f (E, I, R β, k, θ, G, P)f (P G)f (G η), f (P G) models for how the transmission network P depends on the contact network G. f (P G) I{P is possible given G}. In other words, we assume a uniform distribution on all possible transmission trees P given G.

24 The loglikelihood: EIR times observed Parameters: β, k, θ, η Data: E, I, R, (G, P) L(parameters) = f (E, I, R, G, P β, k, θ, η) G,P = G,P f (E, I, R β, k, θ, G, P)f (P G)f (G η), f (G η) is the model for how the contact network G depends on the ERGM parameters η: f (G η) exp{η t g(g)}. Here, the constant of proportionality depends only on η and may be intractable as usual for an ERGM.

25 Priors and updates to parameters Our MCMC-based Bayesian estimation procedure uses prior distributions β gamma θ I, θ E inverse gamma p beta k I, k E gamma The first three of these are conjugate priors; the parameters may be updated using Gibbs sampling. The k I and k E parameters may be updated using a standard Metropolis-Hastings, where proposals are made from a uniform density centered at the current values.

26 Parameter Updates Graph and Tree Parameters Updating the graph (G): Since we are (currently) assuming a dyadic independence graph model, we can update each dyad individually. We calculate the full conditional probability of existence for each possible edge, given all of the other parameters (including P). Updating the transmission tree (P): We must determine, for each affected node except the initial exposed, which node infected it.

27 Notation for Data The exposure, infective, and removal times for node j are denoted by E j, I j, and R j, respectively. Denote the identity of the initial exposed node by κ (which may or may not be known). In order for node a to infect node b, it is necessary that b is exposed during the time that a is infective: I a < E b < R a (1)

28 Example Toy Dataset (slightly less ideal) Node Exposure Time Infective Time Removal Time 1? ? ? ? If necessary... Update each E i individually via Metropolis-Hastings. for i κ: Propose uniformly from possible range for i = κ: No lower bound; use an exponential proposal. The I i are updated in a similarly. To update κ, propose uniformly from children of current κ in P; swap times and direction of transmission.

29 Outline 1 Inference for Contact Networks Epidemic Data 3 Simulation studies 4 Hagelloch Measles Data 5 Future Extensions

30 Exploring the Parameter Space through Simulations The rapid spread of an epidemic throughout a population could be due to either a fast transmission rate (high value of β) or a more fully connected network (large value of p). This can lead to difficulties in estimating these parameters separately. We want to find which areas of the (p β) parameter space lend themselves to meaningful estimation. We simulated Erdős-Rényi networks of 40 individuals with p = 0.1, 0.,..., 1. Over each of these ten networks, we simulated epidemics with five different values of β: 0.01, 0.05, 0.1, 0.5, and 1. Assumed that all E i, I i, and R i times were known.

31 Posterior Scatterplots (true p = 0.) β = 0.01 β = 0.1 β = 1 log(1) log(β) log(0.1) log(0.01) log(0.05) log(0.0) log(0.80) log(0.05) log(0.0) log(0.80) log(p) Scatterplots of the posterior samples for β and p for three different simulations. log(0.05) log(0.0) log(0.80)

32 Sample Epidemics p = 0.1, β = 0.3 p = 1, β = Time Time

33 Sample Epidemics p = 0.1, β = 0.3 p = 1, β = Time Time

34 Sample Epidemics p = 0.1, β = 0.3 p = 1, β = Time Time

35 Sample Epidemics p = 0.1, β = 0.3 p = 1, β = Time Time

36 Sample Epidemics p = 0.1, β = 0.3 p = 1, β = Time Time

37 Outline 1 Inference for Contact Networks Epidemic Data 3 Simulation studies 4 Hagelloch Measles Data 5 Future Extensions

38 Hagelloch Measles Data We consider here an actual data set, namely data from a measles epidemic that spread through the small town of Hagelloch, Germany in The data contains (among other things) proxies for the Infective and Recovery times (we have to infer the Exposure times). All 188 individuals in the susceptible population were infected during the course of the epidemic. (And one outlier.)

39 Hagelloch posterior histograms for p and β Samples from Posterior Distribution of p Samples from Posterior Distribution of β Frequency Frequency p β

40 Results for Inferred Exposure Periods Estimated Posterior Densities for k E! E Estimated Posterior Densities for k E! E Density (0.7) 11.1 (0.8) Density (1.5) 11.8 (.7) Estimated mean (left panel) and variance (right panel) of exposure periods in Hagelloch measles data. Solid lines denote results for all data, and dashed lines indicate results with one outlier removed.

41 Using Information About the Transmission Tree The Hagelloch data set also happens to contain additional information that we can put to use in our inference. In particular, for each infected individual, a putative parent (i.e., the individual most likely to be responsible for the infection) is recorded. We can use this information to form a more informative prior for the transmission tree P. For each node, we can give added prior weight to its putative parent node.

42 Additional Transmission Tree Information Results Uniform Tree Prior 8 x Prior Weight on Putative Parent Node Frequency Frequency Possible Parents for Node Possible Parents for Node 1

43 Posterior distribution for R 0 We use as the notion of R 0 the expected number of first-generation infection events under the model if a randomly chosen node is suddenly infected. R 0 = (Np)P(X < Y ) ( [ ] ) 1 ki = Np 1, 1 + βθ I where X Exponential(β) Y Gamma(k I, θ I ) Frequency Posterior Samples for R0 Conjugate Priors / Uniform Tree Prior R0

44 Posterior Predictive Modeling Number of infectives Day Multiple simulations of epidemics based on draws from the posteriors (observed data in red).

45 Outline 1 Inference for Contact Networks Epidemic Data 3 Simulation studies 4 Hagelloch Measles Data 5 Future Extensions

46 More General ERGM Models One possible extension consists of using a more general ERGM to model the interactions in population. For the general ERGM, the parameter η would replace p. We would have to modify the MCMC algorithm. Unfortunately, for a general ERGM, κ(η) cannot be evaluated in closed form; hence, more complicated updating schemes for η may be necessary. We may be forced to simulate the entire network in order to produce an update, likely using some type of MCMC method.

47 Incorporating Other Types of Data We also want to consider how to best make use of any additional (beyond the Exposure/Infective/Recovery times) data that is available to us. We ve actually already had some success doing this for the Hagelloch data. Additional data (such as viral genetic data) may allow us to partially or fully inform the transmission tree P. This type of genetic data is already used to inform phylogenetic trees; we might consider using similar approaches in order to inform transmission trees.

48 The loglikelihood: Including sequence information S Parameters: β, k, θ, η, µ Data: E, I, R, S, G, P L = f (E, I, R, S, G, P β, k, θ, η, µ) G,P = f (E, I, R, S β, k, θ, G, P)f (S P, E, R, µ)f (P G)f (G η), G,P E 7 time R 6 E 3 R 3 E 6 E 5 E R 5 R R 7 (P, E, R) is the phylogenetic tree shown E and R are the exposure and recovery times; P is the tree. µ are the parameters governing the mutation process (Jukes-Cantor? HKY?)

49 Summary A statistical approach to learning about contact networks from epidemic data requires explicit specification of a contact network model. The model is parametric. In this case, we use an ERGM. Bayesian methods not only provide a means for fitting a complicated model, they also allow incorporation of disease- or network-specific prior information. Estimating parameters allows both simulation of realistic contact networks (e.g., to check model fit) and also understanding of contact processes. This work is still somewhat preliminary in that many extensions are possible; certainly, Erdős-Rényi is not an appropriate model generally.

Bayesian inference for contact networks given epidemic data

Bayesian inference for contact networks given epidemic data Bayesian inference for contact networks given epidemic data Penn State Department of Statistics Technical Report #10-02 Chris Groendyke Department of Statistics Pennsylvania State University cg928@stat.psu.edu

More information

Bayesian Inference for Contact Networks Given Epidemic Data

Bayesian Inference for Contact Networks Given Epidemic Data Bayesian Inference for Contact Networks Given Epidemic Data CHRIS GROENDYKE Department of Statistics, Pennsylvania State University DAVID WELCH Department of Statistics, Pennsylvania State University;

More information

Statistical Inference for Stochastic Epidemic Models

Statistical Inference for Stochastic Epidemic Models Statistical Inference for Stochastic Epidemic Models George Streftaris 1 and Gavin J. Gibson 1 1 Department of Actuarial Mathematics & Statistics, Heriot-Watt University, Riccarton, Edinburgh EH14 4AS,

More information

MCMC 2: Lecture 2 Coding and output. Phil O Neill Theo Kypraios School of Mathematical Sciences University of Nottingham

MCMC 2: Lecture 2 Coding and output. Phil O Neill Theo Kypraios School of Mathematical Sciences University of Nottingham MCMC 2: Lecture 2 Coding and output Phil O Neill Theo Kypraios School of Mathematical Sciences University of Nottingham Contents 1. General (Markov) epidemic model 2. Non-Markov epidemic model 3. Debugging

More information

MCMC 2: Lecture 3 SIR models - more topics. Phil O Neill Theo Kypraios School of Mathematical Sciences University of Nottingham

MCMC 2: Lecture 3 SIR models - more topics. Phil O Neill Theo Kypraios School of Mathematical Sciences University of Nottingham MCMC 2: Lecture 3 SIR models - more topics Phil O Neill Theo Kypraios School of Mathematical Sciences University of Nottingham Contents 1. What can be estimated? 2. Reparameterisation 3. Marginalisation

More information

Algorithmic approaches to fitting ERG models

Algorithmic approaches to fitting ERG models Ruth Hummel, Penn State University Mark Handcock, University of Washington David Hunter, Penn State University Research funded by Office of Naval Research Award No. N00014-08-1-1015 MURI meeting, April

More information

Robust MCMC Algorithms for Bayesian Inference in Stochastic Epidemic Models.

Robust MCMC Algorithms for Bayesian Inference in Stochastic Epidemic Models. Robust MCMC Algorithms for Bayesian Inference in Stochastic Epidemic Models. An Application to the 2001 UK Foot-and-Mouth Outbreak Theodore Kypraios @ University of Nottingham Gareth O. Roberts @ Lancaster

More information

Curved exponential family models for networks

Curved exponential family models for networks Curved exponential family models for networks David R. Hunter, Penn State University Mark S. Handcock, University of Washington February 18, 2005 Available online as Penn State Dept. of Statistics Technical

More information

Bayesian inference and model selection for stochastic epidemics and other coupled hidden Markov models

Bayesian inference and model selection for stochastic epidemics and other coupled hidden Markov models Bayesian inference and model selection for stochastic epidemics and other coupled hidden Markov models (with special attention to epidemics of Escherichia coli O157:H7 in cattle) Simon Spencer 3rd May

More information

STA 4273H: Statistical Machine Learning

STA 4273H: Statistical Machine Learning STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistical Sciences! rsalakhu@cs.toronto.edu! h0p://www.cs.utoronto.ca/~rsalakhu/ Lecture 7 Approximate

More information

Introduction to SEIR Models

Introduction to SEIR Models Department of Epidemiology and Public Health Health Systems Research and Dynamical Modelling Unit Introduction to SEIR Models Nakul Chitnis Workshop on Mathematical Models of Climate Variability, Environmental

More information

Consistency Under Sampling of Exponential Random Graph Models

Consistency Under Sampling of Exponential Random Graph Models Consistency Under Sampling of Exponential Random Graph Models Cosma Shalizi and Alessandro Rinaldo Summary by: Elly Kaizar Remember ERGMs (Exponential Random Graph Models) Exponential family models Sufficient

More information

A Bayesian Approach to Phylogenetics

A Bayesian Approach to Phylogenetics A Bayesian Approach to Phylogenetics Niklas Wahlberg Based largely on slides by Paul Lewis (www.eeb.uconn.edu) An Introduction to Bayesian Phylogenetics Bayesian inference in general Markov chain Monte

More information

Who was Bayes? Bayesian Phylogenetics. What is Bayes Theorem?

Who was Bayes? Bayesian Phylogenetics. What is Bayes Theorem? Who was Bayes? Bayesian Phylogenetics Bret Larget Departments of Botany and of Statistics University of Wisconsin Madison October 6, 2011 The Reverand Thomas Bayes was born in London in 1702. He was the

More information

Bayesian Phylogenetics

Bayesian Phylogenetics Bayesian Phylogenetics Bret Larget Departments of Botany and of Statistics University of Wisconsin Madison October 6, 2011 Bayesian Phylogenetics 1 / 27 Who was Bayes? The Reverand Thomas Bayes was born

More information

Bayesian Methods for Machine Learning

Bayesian Methods for Machine Learning Bayesian Methods for Machine Learning CS 584: Big Data Analytics Material adapted from Radford Neal s tutorial (http://ftp.cs.utoronto.ca/pub/radford/bayes-tut.pdf), Zoubin Ghahramni (http://hunch.net/~coms-4771/zoubin_ghahramani_bayesian_learning.pdf),

More information

Classical and Bayesian inference

Classical and Bayesian inference Classical and Bayesian inference AMS 132 January 18, 2018 Claudia Wehrhahn (UCSC) Classical and Bayesian inference January 18, 2018 1 / 9 Sampling from a Bernoulli Distribution Theorem (Beta-Bernoulli

More information

A network epidemic model for online community commissioning data

A network epidemic model for online community commissioning data Stat Comput () :9 9 https://doi.org/.7/s-7-977- A network epidemic model for online community commissioning data Clement Lee, Andrew Garbett Darren J. Wilkinson Received: 7 February 7 / Accepted: July

More information

Bayesian linear regression

Bayesian linear regression Bayesian linear regression Linear regression is the basis of most statistical modeling. The model is Y i = X T i β + ε i, where Y i is the continuous response X i = (X i1,..., X ip ) T is the corresponding

More information

Bayesian Regression (1/31/13)

Bayesian Regression (1/31/13) STA613/CBB540: Statistical methods in computational biology Bayesian Regression (1/31/13) Lecturer: Barbara Engelhardt Scribe: Amanda Lea 1 Bayesian Paradigm Bayesian methods ask: given that I have observed

More information

Inferring the origin of an epidemic with a dynamic message-passing algorithm

Inferring the origin of an epidemic with a dynamic message-passing algorithm Inferring the origin of an epidemic with a dynamic message-passing algorithm HARSH GUPTA (Based on the original work done by Andrey Y. Lokhov, Marc Mézard, Hiroki Ohta, and Lenka Zdeborová) Paper Andrey

More information

The Monte Carlo Method: Bayesian Networks

The Monte Carlo Method: Bayesian Networks The Method: Bayesian Networks Dieter W. Heermann Methods 2009 Dieter W. Heermann ( Methods)The Method: Bayesian Networks 2009 1 / 18 Outline 1 Bayesian Networks 2 Gene Expression Data 3 Bayesian Networks

More information

DAG models and Markov Chain Monte Carlo methods a short overview

DAG models and Markov Chain Monte Carlo methods a short overview DAG models and Markov Chain Monte Carlo methods a short overview Søren Højsgaard Institute of Genetics and Biotechnology University of Aarhus August 18, 2008 Printed: August 18, 2008 File: DAGMC-Lecture.tex

More information

Efficient Likelihood-Free Inference

Efficient Likelihood-Free Inference Efficient Likelihood-Free Inference Michael Gutmann http://homepages.inf.ed.ac.uk/mgutmann Institute for Adaptive and Neural Computation School of Informatics, University of Edinburgh 8th November 2017

More information

Nonlinear Models. and. Hierarchical Nonlinear Models

Nonlinear Models. and. Hierarchical Nonlinear Models Nonlinear Models and Hierarchical Nonlinear Models Start Simple Progressively Add Complexity Tree Allometries Diameter vs Height with a hierarchical species effect Three response variables: Ht, crown depth,

More information

Variable Selection and Sensitivity Analysis via Dynamic Trees with an application to Computer Code Performance Tuning

Variable Selection and Sensitivity Analysis via Dynamic Trees with an application to Computer Code Performance Tuning Variable Selection and Sensitivity Analysis via Dynamic Trees with an application to Computer Code Performance Tuning Robert B. Gramacy University of Chicago Booth School of Business faculty.chicagobooth.edu/robert.gramacy

More information

Bayesian Approach 2. CSC412 Probabilistic Learning & Reasoning

Bayesian Approach 2. CSC412 Probabilistic Learning & Reasoning CSC412 Probabilistic Learning & Reasoning Lecture 12: Bayesian Parameter Estimation February 27, 2006 Sam Roweis Bayesian Approach 2 The Bayesian programme (after Rev. Thomas Bayes) treats all unnown quantities

More information

PMR Learning as Inference

PMR Learning as Inference Outline PMR Learning as Inference Probabilistic Modelling and Reasoning Amos Storkey Modelling 2 The Exponential Family 3 Bayesian Sets School of Informatics, University of Edinburgh Amos Storkey PMR Learning

More information

F denotes cumulative density. denotes probability density function; (.)

F denotes cumulative density. denotes probability density function; (.) BAYESIAN ANALYSIS: FOREWORDS Notation. System means the real thing and a model is an assumed mathematical form for the system.. he probability model class M contains the set of the all admissible models

More information

Fast Likelihood-Free Inference via Bayesian Optimization

Fast Likelihood-Free Inference via Bayesian Optimization Fast Likelihood-Free Inference via Bayesian Optimization Michael Gutmann https://sites.google.com/site/michaelgutmann University of Helsinki Aalto University Helsinki Institute for Information Technology

More information

Bayesian Statistical Methods. Jeff Gill. Department of Political Science, University of Florida

Bayesian Statistical Methods. Jeff Gill. Department of Political Science, University of Florida Bayesian Statistical Methods Jeff Gill Department of Political Science, University of Florida 234 Anderson Hall, PO Box 117325, Gainesville, FL 32611-7325 Voice: 352-392-0262x272, Fax: 352-392-8127, Email:

More information

CS 343: Artificial Intelligence

CS 343: Artificial Intelligence CS 343: Artificial Intelligence Bayes Nets: Sampling Prof. Scott Niekum The University of Texas at Austin [These slides based on those of Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley.

More information

Delayed Rejection Algorithm to Estimate Bayesian Social Networks

Delayed Rejection Algorithm to Estimate Bayesian Social Networks Dublin Institute of Technology ARROW@DIT Articles School of Mathematics 2014 Delayed Rejection Algorithm to Estimate Bayesian Social Networks Alberto Caimo Dublin Institute of Technology, alberto.caimo@dit.ie

More information

Downloaded from:

Downloaded from: Camacho, A; Kucharski, AJ; Funk, S; Breman, J; Piot, P; Edmunds, WJ (2014) Potential for large outbreaks of Ebola virus disease. Epidemics, 9. pp. 70-8. ISSN 1755-4365 DOI: https://doi.org/10.1016/j.epidem.2014.09.003

More information

Theory of Stochastic Processes 8. Markov chain Monte Carlo

Theory of Stochastic Processes 8. Markov chain Monte Carlo Theory of Stochastic Processes 8. Markov chain Monte Carlo Tomonari Sei sei@mist.i.u-tokyo.ac.jp Department of Mathematical Informatics, University of Tokyo June 8, 2017 http://www.stat.t.u-tokyo.ac.jp/~sei/lec.html

More information

Advanced Machine Learning

Advanced Machine Learning Advanced Machine Learning Nonparametric Bayesian Models --Learning/Reasoning in Open Possible Worlds Eric Xing Lecture 7, August 4, 2009 Reading: Eric Xing Eric Xing @ CMU, 2006-2009 Clustering Eric Xing

More information

STA 4273H: Sta-s-cal Machine Learning

STA 4273H: Sta-s-cal Machine Learning STA 4273H: Sta-s-cal Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistical Sciences! rsalakhu@cs.toronto.edu! h0p://www.cs.utoronto.ca/~rsalakhu/ Lecture 2 In our

More information

Bayesian networks: approximate inference

Bayesian networks: approximate inference Bayesian networks: approximate inference Machine Intelligence Thomas D. Nielsen September 2008 Approximative inference September 2008 1 / 25 Motivation Because of the (worst-case) intractability of exact

More information

Statistical Models for Social Networks with Application to HIV Epidemiology

Statistical Models for Social Networks with Application to HIV Epidemiology Statistical Models for Social Networks with Application to HIV Epidemiology Mark S. Handcock Department of Statistics University of Washington Joint work with Pavel Krivitsky Martina Morris and the U.

More information

Bayesian estimation of complex networks and dynamic choice in the music industry

Bayesian estimation of complex networks and dynamic choice in the music industry Bayesian estimation of complex networks and dynamic choice in the music industry Stefano Nasini Víctor Martínez-de-Albéniz Dept. of Production, Technology and Operations Management, IESE Business School,

More information

MCMC algorithms for fitting Bayesian models

MCMC algorithms for fitting Bayesian models MCMC algorithms for fitting Bayesian models p. 1/1 MCMC algorithms for fitting Bayesian models Sudipto Banerjee sudiptob@biostat.umn.edu University of Minnesota MCMC algorithms for fitting Bayesian models

More information

Introduction to Bayesian Inference

Introduction to Bayesian Inference Introduction to Bayesian Inference p. 1/2 Introduction to Bayesian Inference September 15th, 2010 Reading: Hoff Chapter 1-2 Introduction to Bayesian Inference p. 2/2 Probability: Measurement of Uncertainty

More information

STA 4273H: Statistical Machine Learning

STA 4273H: Statistical Machine Learning STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! http://www.utstat.utoronto.ca/~rsalakhu/ Sidney Smith Hall, Room 6002 Lecture 7 Approximate

More information

Nonparametric Bayesian Methods - Lecture I

Nonparametric Bayesian Methods - Lecture I Nonparametric Bayesian Methods - Lecture I Harry van Zanten Korteweg-de Vries Institute for Mathematics CRiSM Masterclass, April 4-6, 2016 Overview of the lectures I Intro to nonparametric Bayesian statistics

More information

Random Effects Models for Network Data

Random Effects Models for Network Data Random Effects Models for Network Data Peter D. Hoff 1 Working Paper no. 28 Center for Statistics and the Social Sciences University of Washington Seattle, WA 98195-4320 January 14, 2003 1 Department of

More information

ABC random forest for parameter estimation. Jean-Michel Marin

ABC random forest for parameter estimation. Jean-Michel Marin ABC random forest for parameter estimation Jean-Michel Marin Université de Montpellier Institut Montpelliérain Alexander Grothendieck (IMAG) Institut de Biologie Computationnelle (IBC) Labex Numev! joint

More information

GEMF: GENERALIZED EPIDEMIC MODELING FRAMEWORK SOFTWARE IN PYTHON

GEMF: GENERALIZED EPIDEMIC MODELING FRAMEWORK SOFTWARE IN PYTHON GEMF: GENERALIZED EPIDEMIC MODELING FRAMEWORK SOFTWARE IN PYTHON HEMAN SHAKERI Network Science and Engineering Group (NetSE) Department of Electrical and Computer Engineering Kansas State University Manhattan,

More information

The Origin of Deep Learning. Lili Mou Jan, 2015

The Origin of Deep Learning. Lili Mou Jan, 2015 The Origin of Deep Learning Lili Mou Jan, 2015 Acknowledgment Most of the materials come from G. E. Hinton s online course. Outline Introduction Preliminary Boltzmann Machines and RBMs Deep Belief Nets

More information

The Ising model and Markov chain Monte Carlo

The Ising model and Markov chain Monte Carlo The Ising model and Markov chain Monte Carlo Ramesh Sridharan These notes give a short description of the Ising model for images and an introduction to Metropolis-Hastings and Gibbs Markov Chain Monte

More information

Bagging During Markov Chain Monte Carlo for Smoother Predictions

Bagging During Markov Chain Monte Carlo for Smoother Predictions Bagging During Markov Chain Monte Carlo for Smoother Predictions Herbert K. H. Lee University of California, Santa Cruz Abstract: Making good predictions from noisy data is a challenging problem. Methods

More information

Assessing the Goodness-of-Fit of Network Models

Assessing the Goodness-of-Fit of Network Models Assessing the Goodness-of-Fit of Network Models Mark S. Handcock Department of Statistics University of Washington Joint work with David Hunter Steve Goodreau Martina Morris and the U. Washington Network

More information

Sampling Algorithms for Probabilistic Graphical models

Sampling Algorithms for Probabilistic Graphical models Sampling Algorithms for Probabilistic Graphical models Vibhav Gogate University of Washington References: Chapter 12 of Probabilistic Graphical models: Principles and Techniques by Daphne Koller and Nir

More information

Probabilistic Graphical Networks: Definitions and Basic Results

Probabilistic Graphical Networks: Definitions and Basic Results This document gives a cursory overview of Probabilistic Graphical Networks. The material has been gleaned from different sources. I make no claim to original authorship of this material. Bayesian Graphical

More information

CS 188: Artificial Intelligence. Bayes Nets

CS 188: Artificial Intelligence. Bayes Nets CS 188: Artificial Intelligence Probabilistic Inference: Enumeration, Variable Elimination, Sampling Pieter Abbeel UC Berkeley Many slides over this course adapted from Dan Klein, Stuart Russell, Andrew

More information

arxiv: v1 [stat.co] 13 Oct 2017

arxiv: v1 [stat.co] 13 Oct 2017 Bayes factors for partially observed stochastic epidemic models arxiv:1710.04977v1 [stat.co] 13 Oct 2017 Muteb Alharthi Theodore Kypraios Philip D. O Neill June 27, 2018 Abstract We consider the problem

More information

Learning Bayesian Networks for Biomedical Data

Learning Bayesian Networks for Biomedical Data Learning Bayesian Networks for Biomedical Data Faming Liang (Texas A&M University ) Liang, F. and Zhang, J. (2009) Learning Bayesian Networks for Discrete Data. Computational Statistics and Data Analysis,

More information

PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 13: SEQUENTIAL DATA

PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 13: SEQUENTIAL DATA PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 13: SEQUENTIAL DATA Contents in latter part Linear Dynamical Systems What is different from HMM? Kalman filter Its strength and limitation Particle Filter

More information

Markov Chains and MCMC

Markov Chains and MCMC Markov Chains and MCMC CompSci 590.02 Instructor: AshwinMachanavajjhala Lecture 4 : 590.02 Spring 13 1 Recap: Monte Carlo Method If U is a universe of items, and G is a subset satisfying some property,

More information

Bayesian Classification and Regression Trees

Bayesian Classification and Regression Trees Bayesian Classification and Regression Trees James Cussens York Centre for Complex Systems Analysis & Dept of Computer Science University of York, UK 1 Outline Problems for Lessons from Bayesian phylogeny

More information

CPSC 540: Machine Learning

CPSC 540: Machine Learning CPSC 540: Machine Learning More Approximate Inference Mark Schmidt University of British Columbia Winter 2018 Last Time: Approximate Inference We ve been discussing graphical models for density estimation,

More information

CS 361: Probability & Statistics

CS 361: Probability & Statistics March 14, 2018 CS 361: Probability & Statistics Inference The prior From Bayes rule, we know that we can express our function of interest as Likelihood Prior Posterior The right hand side contains the

More information

Supplement to A Hierarchical Approach for Fitting Curves to Response Time Measurements

Supplement to A Hierarchical Approach for Fitting Curves to Response Time Measurements Supplement to A Hierarchical Approach for Fitting Curves to Response Time Measurements Jeffrey N. Rouder Francis Tuerlinckx Paul L. Speckman Jun Lu & Pablo Gomez May 4 008 1 The Weibull regression model

More information

13: Variational inference II

13: Variational inference II 10-708: Probabilistic Graphical Models, Spring 2015 13: Variational inference II Lecturer: Eric P. Xing Scribes: Ronghuo Zheng, Zhiting Hu, Yuntian Deng 1 Introduction We started to talk about variational

More information

Bayesian Quadrature: Model-based Approximate Integration. David Duvenaud University of Cambridge

Bayesian Quadrature: Model-based Approximate Integration. David Duvenaud University of Cambridge Bayesian Quadrature: Model-based Approimate Integration David Duvenaud University of Cambridge The Quadrature Problem ˆ We want to estimate an integral Z = f ()p()d ˆ Most computational problems in inference

More information

STA414/2104 Statistical Methods for Machine Learning II

STA414/2104 Statistical Methods for Machine Learning II STA414/2104 Statistical Methods for Machine Learning II Murat A. Erdogdu & David Duvenaud Department of Computer Science Department of Statistical Sciences Lecture 3 Slide credits: Russ Salakhutdinov Announcements

More information

Bayesian inference for stochastic multitype epidemics in structured populations using sample data

Bayesian inference for stochastic multitype epidemics in structured populations using sample data Bayesian inference for stochastic multitype epidemics in structured populations using sample data PHILIP D. O NEILL School of Mathematical Sciences, University of Nottingham, Nottingham, UK. SUMMARY This

More information

Epidemics in Networks Part 2 Compartmental Disease Models

Epidemics in Networks Part 2 Compartmental Disease Models Epidemics in Networks Part 2 Compartmental Disease Models Joel C. Miller & Tom Hladish 18 20 July 2018 1 / 35 Introduction to Compartmental Models Dynamics R 0 Epidemic Probability Epidemic size Review

More information

Quilting Stochastic Kronecker Graphs to Generate Multiplicative Attribute Graphs

Quilting Stochastic Kronecker Graphs to Generate Multiplicative Attribute Graphs Quilting Stochastic Kronecker Graphs to Generate Multiplicative Attribute Graphs Hyokun Yun (work with S.V.N. Vishwanathan) Department of Statistics Purdue Machine Learning Seminar November 9, 2011 Overview

More information

Generalized Exponential Random Graph Models: Inference for Weighted Graphs

Generalized Exponential Random Graph Models: Inference for Weighted Graphs Generalized Exponential Random Graph Models: Inference for Weighted Graphs James D. Wilson University of North Carolina at Chapel Hill June 18th, 2015 Political Networks, 2015 James D. Wilson GERGMs for

More information

CSC 2541: Bayesian Methods for Machine Learning

CSC 2541: Bayesian Methods for Machine Learning CSC 2541: Bayesian Methods for Machine Learning Radford M. Neal, University of Toronto, 2011 Lecture 4 Problem: Density Estimation We have observed data, y 1,..., y n, drawn independently from some unknown

More information

Introduction to Bayesian Statistics with WinBUGS Part 4 Priors and Hierarchical Models

Introduction to Bayesian Statistics with WinBUGS Part 4 Priors and Hierarchical Models Introduction to Bayesian Statistics with WinBUGS Part 4 Priors and Hierarchical Models Matthew S. Johnson New York ASA Chapter Workshop CUNY Graduate Center New York, NY hspace1in December 17, 2009 December

More information

Approximate Bayesian Computation

Approximate Bayesian Computation Approximate Bayesian Computation Michael Gutmann https://sites.google.com/site/michaelgutmann University of Helsinki and Aalto University 1st December 2015 Content Two parts: 1. The basics of approximate

More information

Comment on A BINOMIAL MOMENT APPROXIMATION SCHEME FOR EPIDEMIC SPREADING IN NETWORKS in U.P.B. Sci. Bull., Series A, Vol. 76, Iss.

Comment on A BINOMIAL MOMENT APPROXIMATION SCHEME FOR EPIDEMIC SPREADING IN NETWORKS in U.P.B. Sci. Bull., Series A, Vol. 76, Iss. Comment on A BINOMIAL MOMENT APPROXIMATION SCHEME FOR EPIDEMIC SPREADING IN NETWORKS in U.P.B. Sci. Bull., Series A, Vol. 76, Iss. 2, 23-3, 24 Istvan Z. Kiss, & Prapanporn Rattana July 2, 24 School of

More information

CPSC 540: Machine Learning

CPSC 540: Machine Learning CPSC 540: Machine Learning MCMC and Non-Parametric Bayes Mark Schmidt University of British Columbia Winter 2016 Admin I went through project proposals: Some of you got a message on Piazza. No news is

More information

27 : Distributed Monte Carlo Markov Chain. 1 Recap of MCMC and Naive Parallel Gibbs Sampling

27 : Distributed Monte Carlo Markov Chain. 1 Recap of MCMC and Naive Parallel Gibbs Sampling 10-708: Probabilistic Graphical Models 10-708, Spring 2014 27 : Distributed Monte Carlo Markov Chain Lecturer: Eric P. Xing Scribes: Pengtao Xie, Khoa Luu In this scribe, we are going to review the Parallel

More information

Introduction to Applied Bayesian Modeling. ICPSR Day 4

Introduction to Applied Bayesian Modeling. ICPSR Day 4 Introduction to Applied Bayesian Modeling ICPSR Day 4 Simple Priors Remember Bayes Law: Where P(A) is the prior probability of A Simple prior Recall the test for disease example where we specified the

More information

1 A simple example. A short introduction to Bayesian statistics, part I Math 217 Probability and Statistics Prof. D.

1 A simple example. A short introduction to Bayesian statistics, part I Math 217 Probability and Statistics Prof. D. probabilities, we ll use Bayes formula. We can easily compute the reverse probabilities A short introduction to Bayesian statistics, part I Math 17 Probability and Statistics Prof. D. Joyce, Fall 014 I

More information

Downscaling Seismic Data to the Meter Scale: Sampling and Marginalization. Subhash Kalla LSU Christopher D. White LSU James S.

Downscaling Seismic Data to the Meter Scale: Sampling and Marginalization. Subhash Kalla LSU Christopher D. White LSU James S. Downscaling Seismic Data to the Meter Scale: Sampling and Marginalization Subhash Kalla LSU Christopher D. White LSU James S. Gunning CSIRO Contents Context of this research Background Data integration

More information

Variational Inference. Sargur Srihari

Variational Inference. Sargur Srihari Variational Inference Sargur srihari@cedar.buffalo.edu 1 Plan of discussion We first describe inference with PGMs and the intractability of exact inference Then give a taxonomy of inference algorithms

More information

Three Disguises of 1 x = e λx

Three Disguises of 1 x = e λx Three Disguises of 1 x = e λx Chathuri Karunarathna Mudiyanselage Rabi K.C. Winfried Just Department of Mathematics, Ohio University Mathematical Biology and Dynamical Systems Seminar Ohio University November

More information

Complex Networks, Course 303A, Spring, Prof. Peter Dodds

Complex Networks, Course 303A, Spring, Prof. Peter Dodds Complex Networks, Course 303A, Spring, 2009 Prof. Peter Dodds Department of Mathematics & Statistics University of Vermont Licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License.

More information

NPFL108 Bayesian inference. Introduction. Filip Jurčíček. Institute of Formal and Applied Linguistics Charles University in Prague Czech Republic

NPFL108 Bayesian inference. Introduction. Filip Jurčíček. Institute of Formal and Applied Linguistics Charles University in Prague Czech Republic NPFL108 Bayesian inference Introduction Filip Jurčíček Institute of Formal and Applied Linguistics Charles University in Prague Czech Republic Home page: http://ufal.mff.cuni.cz/~jurcicek Version: 21/02/2014

More information

STA 294: Stochastic Processes & Bayesian Nonparametrics

STA 294: Stochastic Processes & Bayesian Nonparametrics MARKOV CHAINS AND CONVERGENCE CONCEPTS Markov chains are among the simplest stochastic processes, just one step beyond iid sequences of random variables. Traditionally they ve been used in modelling a

More information

Understanding the contribution of space on the spread of Influenza using an Individual-based model approach

Understanding the contribution of space on the spread of Influenza using an Individual-based model approach Understanding the contribution of space on the spread of Influenza using an Individual-based model approach Shrupa Shah Joint PhD Candidate School of Mathematics and Statistics School of Population and

More information

Approximate inference, Sampling & Variational inference Fall Cours 9 November 25

Approximate inference, Sampling & Variational inference Fall Cours 9 November 25 Approimate inference, Sampling & Variational inference Fall 2015 Cours 9 November 25 Enseignant: Guillaume Obozinski Scribe: Basile Clément, Nathan de Lara 9.1 Approimate inference with MCMC 9.1.1 Gibbs

More information

Chapter 5. Bayesian Statistics

Chapter 5. Bayesian Statistics Chapter 5. Bayesian Statistics Principles of Bayesian Statistics Anything unknown is given a probability distribution, representing degrees of belief [subjective probability]. Degrees of belief [subjective

More information

Deep Poisson Factorization Machines: a factor analysis model for mapping behaviors in journalist ecosystem

Deep Poisson Factorization Machines: a factor analysis model for mapping behaviors in journalist ecosystem 000 001 002 003 004 005 006 007 008 009 010 011 012 013 014 015 016 017 018 019 020 021 022 023 024 025 026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044 045 046 047 048 049 050

More information

An overview of Bayesian analysis

An overview of Bayesian analysis An overview of Bayesian analysis Benjamin Letham Operations Research Center, Massachusetts Institute of Technology, Cambridge, MA bletham@mit.edu May 2012 1 Introduction and Notation This work provides

More information

25 : Graphical induced structured input/output models

25 : Graphical induced structured input/output models 10-708: Probabilistic Graphical Models 10-708, Spring 2013 25 : Graphical induced structured input/output models Lecturer: Eric P. Xing Scribes: Meghana Kshirsagar (mkshirsa), Yiwen Chen (yiwenche) 1 Graph

More information

The Bayesian Choice. Christian P. Robert. From Decision-Theoretic Foundations to Computational Implementation. Second Edition.

The Bayesian Choice. Christian P. Robert. From Decision-Theoretic Foundations to Computational Implementation. Second Edition. Christian P. Robert The Bayesian Choice From Decision-Theoretic Foundations to Computational Implementation Second Edition With 23 Illustrations ^Springer" Contents Preface to the Second Edition Preface

More information

A graph contains a set of nodes (vertices) connected by links (edges or arcs)

A graph contains a set of nodes (vertices) connected by links (edges or arcs) BOLTZMANN MACHINES Generative Models Graphical Models A graph contains a set of nodes (vertices) connected by links (edges or arcs) In a probabilistic graphical model, each node represents a random variable,

More information

Dynamic models. Dependent data The AR(p) model The MA(q) model Hidden Markov models. 6 Dynamic models

Dynamic models. Dependent data The AR(p) model The MA(q) model Hidden Markov models. 6 Dynamic models 6 Dependent data The AR(p) model The MA(q) model Hidden Markov models Dependent data Dependent data Huge portion of real-life data involving dependent datapoints Example (Capture-recapture) capture histories

More information

28 : Approximate Inference - Distributed MCMC

28 : Approximate Inference - Distributed MCMC 10-708: Probabilistic Graphical Models, Spring 2015 28 : Approximate Inference - Distributed MCMC Lecturer: Avinava Dubey Scribes: Hakim Sidahmed, Aman Gupta 1 Introduction For many interesting problems,

More information

CSC 2541: Bayesian Methods for Machine Learning

CSC 2541: Bayesian Methods for Machine Learning CSC 2541: Bayesian Methods for Machine Learning Radford M. Neal, University of Toronto, 2011 Lecture 3 More Markov Chain Monte Carlo Methods The Metropolis algorithm isn t the only way to do MCMC. We ll

More information

Tutorial on Approximate Bayesian Computation

Tutorial on Approximate Bayesian Computation Tutorial on Approximate Bayesian Computation Michael Gutmann https://sites.google.com/site/michaelgutmann University of Helsinki Aalto University Helsinki Institute for Information Technology 16 May 2016

More information

CTDL-Positive Stable Frailty Model

CTDL-Positive Stable Frailty Model CTDL-Positive Stable Frailty Model M. Blagojevic 1, G. MacKenzie 2 1 Department of Mathematics, Keele University, Staffordshire ST5 5BG,UK and 2 Centre of Biostatistics, University of Limerick, Ireland

More information

Some of these slides have been borrowed from Dr. Paul Lewis, Dr. Joe Felsenstein. Thanks!

Some of these slides have been borrowed from Dr. Paul Lewis, Dr. Joe Felsenstein. Thanks! Some of these slides have been borrowed from Dr. Paul Lewis, Dr. Joe Felsenstein. Thanks! Paul has many great tools for teaching phylogenetics at his web site: http://hydrodictyon.eeb.uconn.edu/people/plewis

More information

Bayesian Nonparametrics: Dirichlet Process

Bayesian Nonparametrics: Dirichlet Process Bayesian Nonparametrics: Dirichlet Process Yee Whye Teh Gatsby Computational Neuroscience Unit, UCL http://www.gatsby.ucl.ac.uk/~ywteh/teaching/npbayes2012 Dirichlet Process Cornerstone of modern Bayesian

More information

Bayesian Linear Regression [DRAFT - In Progress]

Bayesian Linear Regression [DRAFT - In Progress] Bayesian Linear Regression [DRAFT - In Progress] David S. Rosenberg Abstract Here we develop some basics of Bayesian linear regression. Most of the calculations for this document come from the basic theory

More information

Bayesian non-parametric model to longitudinally predict churn

Bayesian non-parametric model to longitudinally predict churn Bayesian non-parametric model to longitudinally predict churn Bruno Scarpa Università di Padova Conference of European Statistics Stakeholders Methodologists, Producers and Users of European Statistics

More information