Collapsed Variational Inference for Sum-Product Networks
|
|
- Maria Tyler
- 5 years ago
- Views:
Transcription
1 for Sum-Product Networks Han Zhao 1, Tameem Adel 2, Geoff Gordon 1, Brandon Amos 1 Presented by: Han Zhao Carnegie Mellon University 1, University of Amsterdam 2 June. 20th, / 26
2 Outline Background Sum-Product Networks Variational Inference Collapsed Variational Inference Motivations and Challenges Efficient Marginalization Logarithmic Transformation Experiments Summary 2 / 26
3 Sum-Product Networks Definition A Sum-Product Network (SPN) is a Rooted directed acyclic graph of univariate distributions, sum nodes and product nodes. Value of a product node is the product of its children. Value of a sum node is the weighted sum of its children, where the weights are nonnegative. Value of the network is the value at the root. + w 1 w 2 w 3 I x1 I x1 I x2 I x2 3 / 26
4 Sum-Product Networks Mixture of Trees Each SPN can be decomposed as a mixture of trees: w1 w3 w2 = w 1 +w 2 +w Ix1 I x1 Ix2 I x2 Ix1 Ix2 Ix1 I x2 I x1 I x2 Each tree is a product of univariate distributions. Number of mixture components is Ω(2 Depth ). Each network computes a positive polynomial (posynomial) function of model parameters: τ S n V root (x w) = p t (X i = x i ) w kj t=1 (k,j) T te i=1 4 / 26
5 Sum-Product Networks Bayesian Network Alternatively, each SPN S is equivalent to a Bayesian network B with bipartite structure. H + H1 + H2 H3 + + H4 + H1 H3 H H2 H4 X1 X2 X3 X4 X1 X3 X2 X4 Number of sum nodes in S = Number of hidden variables in B = Θ( S ). B = O(n S ) Number of observable variables in B = Number of variables modeled by S. Typically number of hidden variables number of observable variables. 5 / 26
6 Variational Inference Brief Introduction Bayesian Inference: p(w x) }{{} posterior Often intractable because of: No analytical solution. p(w) }{{} prior Expensive numerical integration. p(x w) }{{} likelihood General idea: find the best approximation in a tractable family of distributions Q: minimize q Q KL[q(w) p(w x)] Typical choice of approximation families: Mean-field, structured mean-field, etc. 6 / 26
7 Variational Inference Brief Introduction Variational method: Optimization-based, deterministic approach for approximate Bayesian inference. inf KL[q(w) p(w x)] sup E q [log p(w, x)] + H[q] q Q q Q Evidence Lower Bound L: log p(x) sup E q [log p(w, x)] + H[q] =: L q Q 7 / 26
8 Motivations and Challenges Bayesian inference algorithms for SPNs: Flexible at incorporating prior knowledge about the structure of SPNs. More robust to overfitting. 8 / 26
9 Motivations and Challenges W 1 W 2 W 3 W m H 1 H 2 H 3 H m X 1 X 2 X 3 X n D W Model parameters, global hidden variables. H Assignments of sum nodes, local hidden variables. X Observable variables. D Number of instances. Challenges for standard VB: Large number of local hidden variables: number of local hidden variables = Number of sum nodes = Θ( S ). 9 / 26
10 Motivations and Challenges W 1 W 2 W 3 W m H 1 H 2 H 3 H m X 1 X 2 X 3 X n D W Model parameters, global hidden variables. H Assignments of sum nodes, local hidden variables. X Observable variables. D Number of instances. Challenges for standard VB: Large number of local hidden variables: number of local hidden variables = Number of sum nodes = Θ( S ). Memory overhead: space complexity O(D S ). 10 / 26
11 Motivations and Challenges W 1 W 2 W 3 W m H 1 H 2 H 3 H m X 1 X 2 X 3 X n D W Model parameters, global hidden variables. H Assignments of sum nodes, local hidden variables. X Observable variables. D Number of instances. Challenges for standard VB: Large number of local hidden variables: number of local hidden variables = Number of sum nodes = Θ( S ). Memory overhead: space complexity O(D S ). Time complexity: O(nD S ). 11 / 26
12 Contributions Our contributions: We obtain better ELBO L to optimize than L, the one obtained by mean-field. Reduced space complexity: O(D S ) O( S ), space complexity is independent of training size. Reduced time complexity: O(nD S ) O(D S ), removing the explicit dependency on the dimension. 12 / 26
13 Efficient Marginalization Recall ELBO in standard VI: L := E q(w,h) [log p(w, h, x)] + H[q(w, h)] Consider the new ELBO in Collapsed VI: L :=E q(w) [log p(w, x)] + H[q(w)] =E q(w) [log h p(w, h, x)] + H[q(w)] We can establish the following inequality: log p(x) L L The new ELBO in Collapsed VI leads to a better lower bound than the one used in standard VI! 13 / 26
14 Comparisons Standard Variational Inference Mean-field assumption: q(w, h) = i q(w i) j q(h j) ELBO: L := E q(w,h) [log p(w, h, x)] + H[q(w, h)] Collapsed Variational Inference for LDA, HDP Collapsed out global hidden variables: q(h) = w q(w, h) dw ELBO: L h := E q(h) [log p(h, x)] + H[q(h)] Better lower bound: L h L Collapsed Variational Inference for SPN Collapsed out local hidden variables: q(w) = h q(w, h) ELBO: L w := E q(w) [log p(w, x)] + H[q(w)] Better lower bound: L w L 14 / 26
15 Efficient Marginalization Time complexity of the exact marginalization incurred in computing h p(w, h, x): Time complexity of marginalization in graphical model G: O(D 2 tw(g) ). Space complexity reduction: 15 / 26
16 Efficient Marginalization Time complexity of the exact marginalization incurred in computing h p(w, h, x): Time complexity of marginalization in graphical model G: O(D 2 tw(g) ). Exact marginalization in BN B with algebraic decision diagram as local factors: O(D B ) = O(nD S ). Space complexity reduction: 16 / 26
17 Efficient Marginalization Time complexity of the exact marginalization incurred in computing h p(w, h, x): Time complexity of marginalization in graphical model G: O(D 2 tw(g) ). Exact marginalization in BN B with algebraic decision diagram as local factors: O(D B ) = O(nD S ). Exact marginalization in SPN S: O(D S ). Space complexity reduction: 17 / 26
18 Efficient Marginalization Time complexity of the exact marginalization incurred in computing h p(w, h, x): Time complexity of marginalization in graphical model G: O(D 2 tw(g) ). Exact marginalization in BN B with algebraic decision diagram as local factors: O(D B ) = O(nD S ). Exact marginalization in SPN S: O(D S ). Space complexity reduction: No posterior over h to approximate anymore. 18 / 26
19 Efficient Marginalization Time complexity of the exact marginalization incurred in computing h p(w, h, x): Time complexity of marginalization in graphical model G: O(D 2 tw(g) ). Exact marginalization in BN B with algebraic decision diagram as local factors: O(D B ) = O(nD S ). Exact marginalization in SPN S: O(D S ). Space complexity reduction: No posterior over h to approximate anymore. No variational variables over h needed: O(D S ) O( S ). 19 / 26
20 Logarithmic Transformation New optimization objective: maximize q Q E q(w) [log h p(w, h, x)] + H[q(w)] which is equivalent to minimize q Q KL[q(w) p(w)] E q(w) [log p(x w)] p(w) prior distribution over w, product of Dirichlets. q(w) variational posterior over w, product of Dirichlets. p(x w) likelihood, not multinomial anymore after marginalization. Non-conjugate q(w) and p(x w), no analytical solution for E q(w) [log p(x w)]. 20 / 26
21 Logarithmic Transformation Key observation: p(x w) = V root (x w) = τ S w kj t=1 (k,j) T te i=1 n p t (X i = x i ) is a posynomial function of w. Make a bijective mapping (change of variable): w = log(w). Dates back to the literature of geometric programming. The new objective after transformation is convex in w. τ S log p(x w) = log exp c t + t=1 (k,j) T te w kj Jensen s inequality to obtain further lower bound. 21 / 26
22 Logarithmic Transformation Further lower bound: E q(w) [log p(x w)] = E q(w )[log p(x w )] log p(x E q (w )[w ]) Relaxed objective: minimize q Q KL[q(w) p(w)] log p(x E }{{} q (w )[w ]) }{{} Regularity Data fitting Roughly, log p(x E q (w )[w ]) corresponds the log-likelihood by setting the weights of SPN as the posterior mean of q(w). Optimized by projected GD. 22 / 26
23 Algorithm Line 4 8 easily parallelizable, distributed version. Sample minibatch in Line 4 8, stochastic version. 23 / 26
24 Experiments Experiments on 20 data sets, report average log-likelihoods, Wilcoxon ranked test. Compared with (O)MLE-SPN and OBMM-SPN. 0 MLE-SPN vs CVB-SPN, Avg. log-likelihoods 0 OMLE-SPN vs OCVB-SPN, Avg. log-likelihoods CVB-Projected GD OCVB-Projected GD MLE-Projected GD OMLE-Projected GD 24 / 26
25 Experiments 0 OBMM vs OCVB, Avg. log-likelihoods 50 OCVB-Projected OBMM-Moment Matching 25 / 26
26 Summary Thanks Q & A 26 / 26
Collapsed Variational Inference for Sum-Product Networks
Han hao HAN.HAO@CS.CMU.EDU Tameem Adel T.M.A.A.HESHAM@UVA.NL Geoff Gordon GGORDON@CS.CMU.EDU Brandon Amos BAMOS@CS.CMU.EDU School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, USA Machine
More informationSum-Product Networks. STAT946 Deep Learning Guest Lecture by Pascal Poupart University of Waterloo October 17, 2017
Sum-Product Networks STAT946 Deep Learning Guest Lecture by Pascal Poupart University of Waterloo October 17, 2017 Introduction Outline What is a Sum-Product Network? Inference Applications In more depth
More informationOn the Relationship between Sum-Product Networks and Bayesian Networks
On the Relationship between Sum-Product Networks and Bayesian Networks International Conference on Machine Learning, 2015 Han Zhao Mazen Melibari Pascal Poupart University of Waterloo, Waterloo, ON, Canada
More information13: 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 informationOnline Algorithms for Sum-Product
Online Algorithms for Sum-Product Networks with Continuous Variables Priyank Jaini Ph.D. Seminar Consistent/Robust Tensor Decomposition And Spectral Learning Offline Bayesian Learning ADF, EP, SGD, oem
More informationStudy Notes on the Latent Dirichlet Allocation
Study Notes on the Latent Dirichlet Allocation Xugang Ye 1. Model Framework A word is an element of dictionary {1,,}. A document is represented by a sequence of words: =(,, ), {1,,}. A corpus is a collection
More informationLinear Time Computation of Moments in Sum-Product Networks
Linear Time Computation of Moments in Sum-Product Netorks Han Zhao Machine Learning Department Carnegie Mellon University Pittsburgh, PA 15213 han.zhao@cs.cmu.edu Geoff Gordon Machine Learning Department
More informationCS839: Probabilistic Graphical Models. Lecture 7: Learning Fully Observed BNs. Theo Rekatsinas
CS839: Probabilistic Graphical Models Lecture 7: Learning Fully Observed BNs Theo Rekatsinas 1 Exponential family: a basic building block For a numeric random variable X p(x ) =h(x)exp T T (x) A( ) = 1
More informationProbabilistic Graphical Models
Probabilistic Graphical Models Lecture 11 CRFs, Exponential Family CS/CNS/EE 155 Andreas Krause Announcements Homework 2 due today Project milestones due next Monday (Nov 9) About half the work should
More informationReadings: K&F: 16.3, 16.4, Graphical Models Carlos Guestrin Carnegie Mellon University October 6 th, 2008
Readings: K&F: 16.3, 16.4, 17.3 Bayesian Param. Learning Bayesian Structure Learning Graphical Models 10708 Carlos Guestrin Carnegie Mellon University October 6 th, 2008 10-708 Carlos Guestrin 2006-2008
More informationProbabilistic Graphical Models
Probabilistic Graphical Models Lecture 9: Variational Inference Relaxations Volkan Cevher, Matthias Seeger Ecole Polytechnique Fédérale de Lausanne 24/10/2011 (EPFL) Graphical Models 24/10/2011 1 / 15
More informationVariational Message Passing. By John Winn, Christopher M. Bishop Presented by Andy Miller
Variational Message Passing By John Winn, Christopher M. Bishop Presented by Andy Miller Overview Background Variational Inference Conjugate-Exponential Models Variational Message Passing Messages Univariate
More informationSTA 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 informationSTA 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 3 Linear
More informationOnline Bayesian Passive-Aggressive Learning
Online Bayesian Passive-Aggressive Learning Full Journal Version: http://qr.net/b1rd Tianlin Shi Jun Zhu ICML 2014 T. Shi, J. Zhu (Tsinghua) BayesPA ICML 2014 1 / 35 Outline Introduction Motivation Framework
More informationMachine 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 information14 : Theory of Variational Inference: Inner and Outer Approximation
10-708: Probabilistic Graphical Models 10-708, Spring 2014 14 : Theory of Variational Inference: Inner and Outer Approximation Lecturer: Eric P. Xing Scribes: Yu-Hsin Kuo, Amos Ng 1 Introduction Last lecture
More informationMachine Learning Techniques for Computer Vision
Machine Learning Techniques for Computer Vision Part 2: Unsupervised Learning Microsoft Research Cambridge x 3 1 0.5 0.2 0 0.5 0.3 0 0.5 1 ECCV 2004, Prague x 2 x 1 Overview of Part 2 Mixture models EM
More informationGraphical Models for Collaborative Filtering
Graphical Models for Collaborative Filtering Le Song Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012 Sequence modeling HMM, Kalman Filter, etc.: Similarity: the same graphical model topology,
More informationVariational Inference (11/04/13)
STA561: Probabilistic machine learning Variational Inference (11/04/13) Lecturer: Barbara Engelhardt Scribes: Matt Dickenson, Alireza Samany, Tracy Schifeling 1 Introduction In this lecture we will further
More informationExponential Families
Exponential Families David M. Blei 1 Introduction We discuss the exponential family, a very flexible family of distributions. Most distributions that you have heard of are in the exponential family. Bernoulli,
More informationAnother Walkthrough of Variational Bayes. Bevan Jones Machine Learning Reading Group Macquarie University
Another Walkthrough of Variational Bayes Bevan Jones Machine Learning Reading Group Macquarie University 2 Variational Bayes? Bayes Bayes Theorem But the integral is intractable! Sampling Gibbs, Metropolis
More informationIntegrated Non-Factorized Variational Inference
Integrated Non-Factorized Variational Inference Shaobo Han, Xuejun Liao and Lawrence Carin Duke University February 27, 2014 S. Han et al. Integrated Non-Factorized Variational Inference February 27, 2014
More informationLecture : Probabilistic Machine Learning
Lecture : Probabilistic Machine Learning Riashat Islam Reasoning and Learning Lab McGill University September 11, 2018 ML : Many Methods with Many Links Modelling Views of Machine Learning Machine Learning
More informationPosterior Regularization
Posterior Regularization 1 Introduction One of the key challenges in probabilistic structured learning, is the intractability of the posterior distribution, for fast inference. There are numerous methods
More informationLecture 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 informationEstimating Latent Variable Graphical Models with Moments and Likelihoods
Estimating Latent Variable Graphical Models with Moments and Likelihoods Arun Tejasvi Chaganty Percy Liang Stanford University June 18, 2014 Chaganty, Liang (Stanford University) Moments and Likelihoods
More informationCOS513 LECTURE 8 STATISTICAL CONCEPTS
COS513 LECTURE 8 STATISTICAL CONCEPTS NIKOLAI SLAVOV AND ANKUR PARIKH 1. MAKING MEANINGFUL STATEMENTS FROM JOINT PROBABILITY DISTRIBUTIONS. A graphical model (GM) represents a family of probability distributions
More informationParametric Models. Dr. Shuang LIANG. School of Software Engineering TongJi University Fall, 2012
Parametric Models Dr. Shuang LIANG School of Software Engineering TongJi University Fall, 2012 Today s Topics Maximum Likelihood Estimation Bayesian Density Estimation Today s Topics Maximum Likelihood
More informationBayesian Machine Learning
Bayesian Machine Learning Andrew Gordon Wilson ORIE 6741 Lecture 2: Bayesian Basics https://people.orie.cornell.edu/andrew/orie6741 Cornell University August 25, 2016 1 / 17 Canonical Machine Learning
More informationTwo Useful Bounds for Variational Inference
Two Useful Bounds for Variational Inference John Paisley Department of Computer Science Princeton University, Princeton, NJ jpaisley@princeton.edu Abstract We review and derive two lower bounds on the
More informationVariational Bayes and Variational Message Passing
Variational Bayes and Variational Message Passing Mohammad Emtiyaz Khan CS,UBC Variational Bayes and Variational Message Passing p.1/16 Variational Inference Find a tractable distribution Q(H) that closely
More informationProbabilistic Graphical Models
Probabilistic Graphical Models Lecture 5 Bayesian Learning of Bayesian Networks CS/CNS/EE 155 Andreas Krause Announcements Recitations: Every Tuesday 4-5:30 in 243 Annenberg Homework 1 out. Due in class
More informationLecture 13 : Variational Inference: Mean Field Approximation
10-708: Probabilistic Graphical Models 10-708, Spring 2017 Lecture 13 : Variational Inference: Mean Field Approximation Lecturer: Willie Neiswanger Scribes: Xupeng Tong, Minxing Liu 1 Problem Setup 1.1
More informationGenerative and Discriminative Approaches to Graphical Models CMSC Topics in AI
Generative and Discriminative Approaches to Graphical Models CMSC 35900 Topics in AI Lecture 2 Yasemin Altun January 26, 2007 Review of Inference on Graphical Models Elimination algorithm finds single
More informationProbabilistic Graphical Models
Probabilistic Graphical Models Brown University CSCI 2950-P, Spring 2013 Prof. Erik Sudderth Lecture 9: Expectation Maximiation (EM) Algorithm, Learning in Undirected Graphical Models Some figures courtesy
More informationIntroduction to Probabilistic Graphical Models: Exercises
Introduction to Probabilistic Graphical Models: Exercises Cédric Archambeau Xerox Research Centre Europe cedric.archambeau@xrce.xerox.com Pascal Bootcamp Marseille, France, July 2010 Exercise 1: basics
More informationBayesian Machine Learning
Bayesian Machine Learning Andrew Gordon Wilson ORIE 6741 Lecture 4 Occam s Razor, Model Construction, and Directed Graphical Models https://people.orie.cornell.edu/andrew/orie6741 Cornell University September
More information13 : Variational Inference: Loopy Belief Propagation and Mean Field
10-708: Probabilistic Graphical Models 10-708, Spring 2012 13 : Variational Inference: Loopy Belief Propagation and Mean Field Lecturer: Eric P. Xing Scribes: Peter Schulam and William Wang 1 Introduction
More informationGraphical Models and Kernel Methods
Graphical Models and Kernel Methods Jerry Zhu Department of Computer Sciences University of Wisconsin Madison, USA MLSS June 17, 2014 1 / 123 Outline Graphical Models Probabilistic Inference Directed vs.
More informationProbabilistic Graphical Models
Probabilistic Graphical Models Lecture 12 Dynamical Models CS/CNS/EE 155 Andreas Krause Homework 3 out tonight Start early!! Announcements Project milestones due today Please email to TAs 2 Parameter learning
More informationNote 1: Varitional Methods for Latent Dirichlet Allocation
Technical Note Series Spring 2013 Note 1: Varitional Methods for Latent Dirichlet Allocation Version 1.0 Wayne Xin Zhao batmanfly@gmail.com Disclaimer: The focus of this note was to reorganie the content
More informationLecture 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 informationChapter 16. Structured Probabilistic Models for Deep Learning
Peng et al.: Deep Learning and Practice 1 Chapter 16 Structured Probabilistic Models for Deep Learning Peng et al.: Deep Learning and Practice 2 Structured Probabilistic Models way of using graphs to describe
More informationLatent Dirichlet Allocation
Outlines Advanced Artificial Intelligence October 1, 2009 Outlines Part I: Theoretical Background Part II: Application and Results 1 Motive Previous Research Exchangeability 2 Notation and Terminology
More informationCS6220: DATA MINING TECHNIQUES
CS6220: DATA MINING TECHNIQUES Matrix Data: Clustering: Part 2 Instructor: Yizhou Sun yzsun@ccs.neu.edu October 19, 2014 Methods to Learn Matrix Data Set Data Sequence Data Time Series Graph & Network
More informationOverfitting, Bias / Variance Analysis
Overfitting, Bias / Variance Analysis Professor Ameet Talwalkar Professor Ameet Talwalkar CS260 Machine Learning Algorithms February 8, 207 / 40 Outline Administration 2 Review of last lecture 3 Basic
More informationIntroduction to Machine Learning
Introduction to Machine Learning Logistic Regression Varun Chandola Computer Science & Engineering State University of New York at Buffalo Buffalo, NY, USA chandola@buffalo.edu Chandola@UB CSE 474/574
More informationVariational Scoring of Graphical Model Structures
Variational Scoring of Graphical Model Structures Matthew J. Beal Work with Zoubin Ghahramani & Carl Rasmussen, Toronto. 15th September 2003 Overview Bayesian model selection Approximations using Variational
More informationProbabilistic Graphical Models: MRFs and CRFs. CSE628: Natural Language Processing Guest Lecturer: Veselin Stoyanov
Probabilistic Graphical Models: MRFs and CRFs CSE628: Natural Language Processing Guest Lecturer: Veselin Stoyanov Why PGMs? PGMs can model joint probabilities of many events. many techniques commonly
More informationBayesian 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 informationLearning Bayesian network : Given structure and completely observed data
Learning Bayesian network : Given structure and completely observed data Probabilistic Graphical Models Sharif University of Technology Spring 2017 Soleymani Learning problem Target: true distribution
More informationBasic math for biology
Basic math for biology Lei Li Florida State University, Feb 6, 2002 The EM algorithm: setup Parametric models: {P θ }. Data: full data (Y, X); partial data Y. Missing data: X. Likelihood and maximum likelihood
More informationp L yi z n m x N n xi
y i z n x n N x i Overview Directed and undirected graphs Conditional independence Exact inference Latent variables and EM Variational inference Books statistical perspective Graphical Models, S. Lauritzen
More informationIntroduction to Probabilistic Machine Learning
Introduction to Probabilistic Machine Learning Piyush Rai Dept. of CSE, IIT Kanpur (Mini-course 1) Nov 03, 2015 Piyush Rai (IIT Kanpur) Introduction to Probabilistic Machine Learning 1 Machine Learning
More informationExpectation Propagation Algorithm
Expectation Propagation Algorithm 1 Shuang Wang School of Electrical and Computer Engineering University of Oklahoma, Tulsa, OK, 74135 Email: {shuangwang}@ou.edu This note contains three parts. First,
More informationPROBABILITY DISTRIBUTIONS. J. Elder CSE 6390/PSYC 6225 Computational Modeling of Visual Perception
PROBABILITY DISTRIBUTIONS Credits 2 These slides were sourced and/or modified from: Christopher Bishop, Microsoft UK Parametric Distributions 3 Basic building blocks: Need to determine given Representation:
More informationNon-Parametric Bayes
Non-Parametric Bayes Mark Schmidt UBC Machine Learning Reading Group January 2016 Current Hot Topics in Machine Learning Bayesian learning includes: Gaussian processes. Approximate inference. Bayesian
More informationLatent Dirichlet Allocation (LDA)
Latent Dirichlet Allocation (LDA) D. Blei, A. Ng, and M. Jordan. Journal of Machine Learning Research, 3:993-1022, January 2003. Following slides borrowed ant then heavily modified from: Jonathan Huang
More informationOnline Bayesian Passive-Agressive Learning
Online Bayesian Passive-Agressive Learning International Conference on Machine Learning, 2014 Tianlin Shi Jun Zhu Tsinghua University, China 21 August 2015 Presented by: Kyle Ulrich Introduction Online
More informationECE521 Tutorial 11. Topic Review. ECE521 Winter Credits to Alireza Makhzani, Alex Schwing, Rich Zemel and TAs for slides. ECE521 Tutorial 11 / 4
ECE52 Tutorial Topic Review ECE52 Winter 206 Credits to Alireza Makhzani, Alex Schwing, Rich Zemel and TAs for slides ECE52 Tutorial ECE52 Winter 206 Credits to Alireza / 4 Outline K-means, PCA 2 Bayesian
More informationChris Bishop s PRML Ch. 8: Graphical Models
Chris Bishop s PRML Ch. 8: Graphical Models January 24, 2008 Introduction Visualize the structure of a probabilistic model Design and motivate new models Insights into the model s properties, in particular
More informationLecture 8: Graphical models for Text
Lecture 8: Graphical models for Text 4F13: Machine Learning Joaquin Quiñonero-Candela and Carl Edward Rasmussen Department of Engineering University of Cambridge http://mlg.eng.cam.ac.uk/teaching/4f13/
More informationOnline and Distributed Bayesian Moment Matching for Parameter Learning in Sum-Product Networks
Online and Distributed Bayesian Moment Matching for Parameter Learning in Sum-Product Networks Abdullah Rashwan Han Zhao Pascal Poupart Computer Science University of Waterloo arashwan@uwaterloo.ca Machine
More information11. Learning graphical models
Learning graphical models 11-1 11. Learning graphical models Maximum likelihood Parameter learning Structural learning Learning partially observed graphical models Learning graphical models 11-2 statistical
More informationIntroduction to Machine Learning
Introduction to Machine Learning Linear Regression Varun Chandola Computer Science & Engineering State University of New York at Buffalo Buffalo, NY, USA chandola@buffalo.edu Chandola@UB CSE 474/574 1
More informationOutline Lecture 2 2(32)
Outline Lecture (3), Lecture Linear Regression and Classification it is our firm belief that an understanding of linear models is essential for understanding nonlinear ones Thomas Schön Division of Automatic
More informationEM & Variational Bayes
EM & Variational Bayes Hanxiao Liu September 9, 2014 1 / 19 Outline 1. EM Algorithm 1.1 Introduction 1.2 Example: Mixture of vmfs 2. Variational Bayes 2.1 Introduction 2.2 Example: Bayesian Mixture of
More informationECE521 week 3: 23/26 January 2017
ECE521 week 3: 23/26 January 2017 Outline Probabilistic interpretation of linear regression - Maximum likelihood estimation (MLE) - Maximum a posteriori (MAP) estimation Bias-variance trade-off Linear
More informationStat 5101 Lecture Notes
Stat 5101 Lecture Notes Charles J. Geyer Copyright 1998, 1999, 2000, 2001 by Charles J. Geyer May 7, 2001 ii Stat 5101 (Geyer) Course Notes Contents 1 Random Variables and Change of Variables 1 1.1 Random
More informationIntroduction to Machine Learning Midterm Exam Solutions
10-701 Introduction to Machine Learning Midterm Exam Solutions Instructors: Eric Xing, Ziv Bar-Joseph 17 November, 2015 There are 11 questions, for a total of 100 points. This exam is open book, open notes,
More informationGaussian Process Regression
Gaussian Process Regression 4F1 Pattern Recognition, 21 Carl Edward Rasmussen Department of Engineering, University of Cambridge November 11th - 16th, 21 Rasmussen (Engineering, Cambridge) Gaussian Process
More informationLecture 21: Spectral Learning for Graphical Models
10-708: Probabilistic Graphical Models 10-708, Spring 2016 Lecture 21: Spectral Learning for Graphical Models Lecturer: Eric P. Xing Scribes: Maruan Al-Shedivat, Wei-Cheng Chang, Frederick Liu 1 Motivation
More informationQuantitative Biology II Lecture 4: Variational Methods
10 th March 2015 Quantitative Biology II Lecture 4: Variational Methods Gurinder Singh Mickey Atwal Center for Quantitative Biology Cold Spring Harbor Laboratory Image credit: Mike West Summary Approximate
More informationIntroduction to Bayesian inference
Introduction to Bayesian inference Thomas Alexander Brouwer University of Cambridge tab43@cam.ac.uk 17 November 2015 Probabilistic models Describe how data was generated using probability distributions
More informationIntroduction 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 informationThe Expectation-Maximization Algorithm
1/29 EM & Latent Variable Models Gaussian Mixture Models EM Theory The Expectation-Maximization Algorithm Mihaela van der Schaar Department of Engineering Science University of Oxford MLE for Latent Variable
More informationBayesian Learning in Undirected Graphical Models
Bayesian Learning in Undirected Graphical Models Zoubin Ghahramani Gatsby Computational Neuroscience Unit University College London, UK http://www.gatsby.ucl.ac.uk/ Work with: Iain Murray and Hyun-Chul
More informationGraphical Models 359
8 Graphical Models Probabilities play a central role in modern pattern recognition. We have seen in Chapter 1 that probability theory can be expressed in terms of two simple equations corresponding to
More informationDEPARTMENT OF COMPUTER SCIENCE Autumn Semester MACHINE LEARNING AND ADAPTIVE INTELLIGENCE
Data Provided: None DEPARTMENT OF COMPUTER SCIENCE Autumn Semester 203 204 MACHINE LEARNING AND ADAPTIVE INTELLIGENCE 2 hours Answer THREE of the four questions. All questions carry equal weight. Figures
More informationσ(a) = a N (x; 0, 1 2 ) dx. σ(a) = Φ(a) =
Until now we have always worked with likelihoods and prior distributions that were conjugate to each other, allowing the computation of the posterior distribution to be done in closed form. Unfortunately,
More informationCollapsed Variational Bayesian Inference for Hidden Markov Models
Collapsed Variational Bayesian Inference for Hidden Markov Models Pengyu Wang, Phil Blunsom Department of Computer Science, University of Oxford International Conference on Artificial Intelligence and
More informationStructure Learning: the good, the bad, the ugly
Readings: K&F: 15.1, 15.2, 15.3, 15.4, 15.5 Structure Learning: the good, the bad, the ugly Graphical Models 10708 Carlos Guestrin Carnegie Mellon University September 29 th, 2006 1 Understanding the uniform
More informationVariational Learning : From exponential families to multilinear systems
Variational Learning : From exponential families to multilinear systems Ananth Ranganathan th February 005 Abstract This note aims to give a general overview of variational inference on graphical models.
More informationVariational Principal Components
Variational Principal Components Christopher M. Bishop Microsoft Research 7 J. J. Thomson Avenue, Cambridge, CB3 0FB, U.K. cmbishop@microsoft.com http://research.microsoft.com/ cmbishop In Proceedings
More informationCSci 8980: Advanced Topics in Graphical Models Gaussian Processes
CSci 8980: Advanced Topics in Graphical Models Gaussian Processes Instructor: Arindam Banerjee November 15, 2007 Gaussian Processes Outline Gaussian Processes Outline Parametric Bayesian Regression Gaussian
More informationMark your answers ON THE EXAM ITSELF. If you are not sure of your answer you may wish to provide a brief explanation.
CS 189 Spring 2015 Introduction to Machine Learning Midterm You have 80 minutes for the exam. The exam is closed book, closed notes except your one-page crib sheet. No calculators or electronic items.
More informationIntroduction to Machine Learning Midterm, Tues April 8
Introduction to Machine Learning 10-701 Midterm, Tues April 8 [1 point] Name: Andrew ID: Instructions: You are allowed a (two-sided) sheet of notes. Exam ends at 2:45pm Take a deep breath and don t spend
More informationPATTERN RECOGNITION AND MACHINE LEARNING
PATTERN RECOGNITION AND MACHINE LEARNING Chapter 1. Introduction Shuai Huang April 21, 2014 Outline 1 What is Machine Learning? 2 Curve Fitting 3 Probability Theory 4 Model Selection 5 The curse of dimensionality
More informationA 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 informationPatterns of Scalable Bayesian Inference Background (Session 1)
Patterns of Scalable Bayesian Inference Background (Session 1) Jerónimo Arenas-García Universidad Carlos III de Madrid jeronimo.arenas@gmail.com June 14, 2017 1 / 15 Motivation. Bayesian Learning principles
More informationBayesian Modeling of Conditional Distributions
Bayesian Modeling of Conditional Distributions John Geweke University of Iowa Indiana University Department of Economics February 27, 2007 Outline Motivation Model description Methods of inference Earnings
More informationBayesian Learning. HT2015: SC4 Statistical Data Mining and Machine Learning. Maximum Likelihood Principle. The Bayesian Learning Framework
HT5: SC4 Statistical Data Mining and Machine Learning Dino Sejdinovic Department of Statistics Oxford http://www.stats.ox.ac.uk/~sejdinov/sdmml.html Maximum Likelihood Principle A generative model for
More informationBayesian Analysis for Natural Language Processing Lecture 2
Bayesian Analysis for Natural Language Processing Lecture 2 Shay Cohen February 4, 2013 Administrativia The class has a mailing list: coms-e6998-11@cs.columbia.edu Need two volunteers for leading a discussion
More informationRecent Advances in Bayesian Inference Techniques
Recent Advances in Bayesian Inference Techniques Christopher M. Bishop Microsoft Research, Cambridge, U.K. research.microsoft.com/~cmbishop SIAM Conference on Data Mining, April 2004 Abstract Bayesian
More informationThe 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 informationLearning P-maps Param. Learning
Readings: K&F: 3.3, 3.4, 16.1, 16.2, 16.3, 16.4 Learning P-maps Param. Learning Graphical Models 10708 Carlos Guestrin Carnegie Mellon University September 24 th, 2008 10-708 Carlos Guestrin 2006-2008
More informationThe 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 informationText Mining for Economics and Finance Latent Dirichlet Allocation
Text Mining for Economics and Finance Latent Dirichlet Allocation Stephen Hansen Text Mining Lecture 5 1 / 45 Introduction Recall we are interested in mixed-membership modeling, but that the plsi model
More informationVariational Mixture of Gaussians. Sargur Srihari
Variational Mixture of Gaussians Sargur srihari@cedar.buffalo.edu 1 Objective Apply variational inference machinery to Gaussian Mixture Models Demonstrates how Bayesian treatment elegantly resolves difficulties
More information