THE POWER OF CERTAINTY

Size: px
Start display at page:

Download "THE POWER OF CERTAINTY"

Transcription

1 A DIRICHLET-MULTINOMIAL APPROACH TO BELIEF PROPAGATION Dhivya Eswaran* CMU Stephan Guennemann TUM Christos Faloutsos CMU

2 PROBLEM MOTIVATION NETCONF GUARANTEES PROBLEM

3 RECOMMENDATION BOB JOHN BOB ALICE CAROL SMITH ALICE CAROL SMITH ADS PLACEMENT

4 GRAPH LABELING / NODE CLASSIFICATION H GIVEN a graph of nodes & edges SMITH labels for a few nodes label compatibility ALICE BOB FIND labels of all nodes H JOHN CAROL

5 NODE CLASSIFICATION IS HARD 4 x 30 x 1 x 15 x SMITH JOHN Higher fraction of android friends Who is more likely to buy an Android phone? Higher number of android friends 5

6 PROBLEM MOTIVATION NETCONF GUARANTEES MOTIVATION

7 CLASSIFICATION BY PROPAGATION INITIALIZE: Set nodes to random/given values. BOB PROPAGATE: Update each node s value based on the values of its neighbors. ALICE SMITH CAROL CONVERGENCE: If no value changes, terminate. Else continue propagation. Q1. What are values here? Q2. How are they updated? 7

8 BELIEF PROPAGATION [0.5, 0.5] DETAILS! A1. Values : beliefs (probability vectors) [0.4, 0.6] BOB [0.8, 0.2] (i) A2. Update : 2 stages Each neighbor sends a message m vu (i) kx H(i, j)e v (j) Y m wu (i) ALICE CAROL j=1 v2n (v)\u SMITH [0.73, 0.27] b u (i) e u (i) (ii) Node updates its belief based on messages Y v2n (u) m vu (i) 8

9 BP LEADS TO COUNTER-INTUITIVE RESULTS Who is more likely to buy an Android phone? 13 x 110 x SMITH 3 x 100 x JOHN BP INTUITION 9

10 MAIN IDEA belief distributions as values probability certainty / confidence (absolute count of neighbors) belief / leaning (ratio of android : apple neighbors) 10

11 PROBLEM MOTIVATION NETCONF GUARANTEES NETCONF

12 NODE CLASSIFICATION WITH CERTAINTY GIVEN a graph of nodes & edges belief distributions for a few nodes H label compatibility FIND SMITH belief distributions of all nodes SUBJECT TO ALICE BOB theoretically-grounded algorithm JOHN fast & scalable implementation CAROL 12

13 DIRICHLET BELIEF DISTRIBUTIONS (1/2) BACKGROUND 2D Dirichlet distribution (Beta distribution) p(x; +1, + 1) / x (1 x) Belief/Leaning : # # 13

14 DIRICHLET BELIEF DISTRIBUTIONS (2/2) BACKGROUND Certainty: + EXTENDS TO ANY NUMBER OF DIMENSIONS!! Example: 3 dimensions (talkative / silent / confused) Image source: UBC Wiki 14

15 MULTINOMIAL MESSAGE DISTRIBUTIONS DETAILS! BP belief update rule b u (i) e u (i) Y v2n (u) m vu (i) DIRICHLET DISTRIBUTION MULTINOMIAL DISTRIBUTION! NETCONF belief update rule b u ĕ u + X v2n (u) m vu parameters of belief distribution parameters of message distribution 15

16 MESSAGES FROM BELIEFS (a, b) SMITH (a, b) JOHN perfect homophily no network effects perfect heterophily (0, 0) (b, a) M = k k 1 H 1 k + 16

17 NETCONF BELIEF & MESSAGE UPDATE RULES message update m vu M u + X v2n (v)\u m wu 1 A belief update b u ĕ u + X v2n (u) m vu M b u m uv 17

18 NETCONF MATRIX BELIEF UPDATE B Ĕ +(A BM D BM 2 )(I M 2 ) 1 final belief distribution prior belief distribution adjacency diagonal degree modulation ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 18

19 PROBLEM MOTIVATION NETCONF GUARANTEES GUARANTEES

20 KEY THEORETICAL QUESTIONS UNIQUENESS Is the steady state solution unique? CONVERGENCE Can we predict convergence? 20

21 NETCONF HAS CLOSED-FORM SOLUTION DETAILS! ITERATIVE UPDATE B Ĕ +(A BM D BM 2 )(I M 2 ) 1 Roth s Column Lemma CLOSED FORM vec( B) = D 1 I (M ˆM) T A +(M 2 ˆM) T vec( Ĕ) 21

22 GUARANTEES PRECISE CONVERGENCE GUARANTEES CLOSED FORM vec( B) = I (M ˆM) T A +(M 2 ˆM) T D 1 vec( Ĕ) DETAILS! NECESSARY & SUFFICIENT CONDITION (M ˆM) T A +(M 2 ˆM) T D < 1 GRAPH STRUCTURE LABEL COMPATIBILITY 22

23 KEY THEORETICAL GUARANTEES CLOSED-FORM Closed-form solution and unique fixed point! CONVERGENCE Necessary and sufficient conditions for convergence! 23

24 PROBLEM MOTIVATION NETCONF GUARANTEES

25 KEY QUESTIONS FOR EFFECTIVENESS Improves accuracy & precision? SCALABILITY Fast and scalable? INTERPRETABILITY Are final results interpretable? 25

26 DATA POLBLOGS DBLP POKEC 1.5K nodes, 19K edges 28K nodes, 67K edges 1.6M nodes, 30M edges 2 classes 4 classes 10 classes 26

27 NETCONF IS ACCURATE AND PRECISE HIGHER ACCURACY % BETTER PRECISION Ideal DATASET BP NETCONF POLBLOGS DBLP POKEC

28 NETCONF IS FAST AND SCALABLE 30M edges in ~7 seconds! Linear scaling with graph size 28

29 NETCONF GIVES INTERPRETABLE RESULTS TOP DATABASES AUTHORS IN DBLP AUTHOR H-index AUTHOR H-index Michael J Carey 48 Jiawei Han 139 Rakesh Agrawal 96 Annie W Shum - Jiawei Han 139 Werner Keibling - Hamid Pirahesh 40 Xiaofang Zhou 36 David J Dewitt 81 Bertram Ludascher 45 Serge Abiteboul 77 Amarnath Gupta - NETCONF Many papers & high H1 indices! BP 29

30 KEY EXPERIMENTAL FINDINGS EFFECTIVENESS Improves accuracy & precision! SCALABILITY Scales linearly with graph size! INTERPRETABILITY Certainty scores reflect intuition! 30

31 SUMMARY SUMMARY: NETCONF Theoretically grounded Closed-form solution Convergence guarantees Improved performance Fast and scalable Interpretable results Questions? 31

BEYOND ASSORTATIVITY PROCLIVITY INDEX FOR ATTRIBUTED NETWORKS (PRONE) Reihaneh Rabbany Dhivya Eswaran*

BEYOND ASSORTATIVITY PROCLIVITY INDEX FOR ATTRIBUTED NETWORKS (PRONE) Reihaneh Rabbany Dhivya Eswaran* PROCLIVITY INDEX FOR ATTRIBUTED NETWORKS (PRONE) Reihaneh Rabbany rrabbany@andrew.cmu.edu Artur W. Dubrawski awd@andrew.cmu.edu Dhivya Eswaran* deswaran@cs.cmu.edu Christos Faloutsos christos@cs.cmu.edu

More information

Node similarity and classification

Node similarity and classification Node similarity and classification Davide Mottin, Anton Tsitsulin HassoPlattner Institute Graph Mining course Winter Semester 2017 Acknowledgements Some part of this lecture is taken from: http://web.eecs.umich.edu/~dkoutra/tut/icdm14.html

More information

Sparse representation classification and positive L1 minimization

Sparse representation classification and positive L1 minimization Sparse representation classification and positive L1 minimization Cencheng Shen Joint Work with Li Chen, Carey E. Priebe Applied Mathematics and Statistics Johns Hopkins University, August 5, 2014 Cencheng

More information

Distributed Systems. 06. Logical clocks. Paul Krzyzanowski. Rutgers University. Fall 2017

Distributed Systems. 06. Logical clocks. Paul Krzyzanowski. Rutgers University. Fall 2017 Distributed Systems 06. Logical clocks Paul Krzyzanowski Rutgers University Fall 2017 2014-2017 Paul Krzyzanowski 1 Logical clocks Assign sequence numbers to messages All cooperating processes can agree

More information

9 Forward-backward algorithm, sum-product on factor graphs

9 Forward-backward algorithm, sum-product on factor graphs Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science 6.438 Algorithms For Inference Fall 2014 9 Forward-backward algorithm, sum-product on factor graphs The previous

More information

Loopy Belief Propagation for Bipartite Maximum Weight b-matching

Loopy Belief Propagation for Bipartite Maximum Weight b-matching Loopy Belief Propagation for Bipartite Maximum Weight b-matching Bert Huang and Tony Jebara Computer Science Department Columbia University New York, NY 10027 Outline 1. Bipartite Weighted b-matching 2.

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 11 Project

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

Computer Vision Group Prof. Daniel Cremers. 14. Clustering

Computer Vision Group Prof. Daniel Cremers. 14. Clustering Group Prof. Daniel Cremers 14. Clustering Motivation Supervised learning is good for interaction with humans, but labels from a supervisor are hard to obtain Clustering is unsupervised learning, i.e. it

More information

Latent Dirichlet Allocation

Latent Dirichlet Allocation Latent Dirichlet Allocation 1 Directed Graphical Models William W. Cohen Machine Learning 10-601 2 DGMs: The Burglar Alarm example Node ~ random variable Burglar Earthquake Arcs define form of probability

More information

Machine Learning for Data Science (CS4786) Lecture 24

Machine Learning for Data Science (CS4786) Lecture 24 Machine Learning for Data Science (CS4786) Lecture 24 Graphical Models: Approximate Inference Course Webpage : http://www.cs.cornell.edu/courses/cs4786/2016sp/ BELIEF PROPAGATION OR MESSAGE PASSING Each

More information

Probabilistic Graphical Models

Probabilistic Graphical Models Probabilistic Graphical Models Brown University CSCI 295-P, Spring 213 Prof. Erik Sudderth Lecture 11: Inference & Learning Overview, Gaussian Graphical Models Some figures courtesy Michael Jordan s draft

More information

Linear Classification: Perceptron

Linear Classification: Perceptron Linear Classification: Perceptron Yufei Tao Department of Computer Science and Engineering Chinese University of Hong Kong 1 / 18 Y Tao Linear Classification: Perceptron In this lecture, we will consider

More information

Expectation Propagation in Factor Graphs: A Tutorial

Expectation Propagation in Factor Graphs: A Tutorial DRAFT: Version 0.1, 28 October 2005. Do not distribute. Expectation Propagation in Factor Graphs: A Tutorial Charles Sutton October 28, 2005 Abstract Expectation propagation is an important variational

More information

Probabilistic Graphical Models

Probabilistic Graphical Models Probabilistic Graphical Models Brown University CSCI 2950-P, Spring 2013 Prof. Erik Sudderth Lecture 12: Gaussian Belief Propagation, State Space Models and Kalman Filters Guest Kalman Filter Lecture by

More information

Running Time. Assumption. All capacities are integers between 1 and C.

Running Time. Assumption. All capacities are integers between 1 and C. Running Time Assumption. All capacities are integers between and. Invariant. Every flow value f(e) and every residual capacities c f (e) remains an integer throughout the algorithm. Theorem. The algorithm

More information

On Top-k Structural. Similarity Search. Pei Lee, Laks V.S. Lakshmanan University of British Columbia Vancouver, BC, Canada

On Top-k Structural. Similarity Search. Pei Lee, Laks V.S. Lakshmanan University of British Columbia Vancouver, BC, Canada On Top-k Structural 1 Similarity Search Pei Lee, Laks V.S. Lakshmanan University of British Columbia Vancouver, BC, Canada Jeffrey Xu Yu Chinese University of Hong Kong Hong Kong, China 2014/10/14 Pei

More information

Lecture 18 Generalized Belief Propagation and Free Energy Approximations

Lecture 18 Generalized Belief Propagation and Free Energy Approximations Lecture 18, Generalized Belief Propagation and Free Energy Approximations 1 Lecture 18 Generalized Belief Propagation and Free Energy Approximations In this lecture we talked about graphical models and

More information

Generative Models for Discrete Data

Generative Models for Discrete Data Generative Models for Discrete Data ddebarr@uw.edu 2016-04-21 Agenda Bayesian Concept Learning Beta-Binomial Model Dirichlet-Multinomial Model Naïve Bayes Classifiers Bayesian Concept Learning Numbers

More information

Mining Newsgroups Using Networks Arising From Social Behavior by Rakesh Agrawal et al. Presented by Will Lee

Mining Newsgroups Using Networks Arising From Social Behavior by Rakesh Agrawal et al. Presented by Will Lee Mining Newsgroups Using Networks Arising From Social Behavior by Rakesh Agrawal et al. Presented by Will Lee wwlee1@uiuc.edu September 28, 2004 Motivation IR on newsgroups is challenging due to lack of

More information

SVD, Power method, and Planted Graph problems (+ eigenvalues of random matrices)

SVD, Power method, and Planted Graph problems (+ eigenvalues of random matrices) Chapter 14 SVD, Power method, and Planted Graph problems (+ eigenvalues of random matrices) Today we continue the topic of low-dimensional approximation to datasets and matrices. Last time we saw the singular

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 3 Linear

More information

STA 414/2104: Machine Learning

STA 414/2104: Machine Learning STA 414/2104: Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistics! rsalakhu@cs.toronto.edu! http://www.cs.toronto.edu/~rsalakhu/ Lecture 9 Sequential Data So far

More information

Decomposition Methods for Large Scale LP Decoding

Decomposition Methods for Large Scale LP Decoding Decomposition Methods for Large Scale LP Decoding Siddharth Barman Joint work with Xishuo Liu, Stark Draper, and Ben Recht Outline Background and Problem Setup LP Decoding Formulation Optimization Framework

More information

Linear Programming Duality

Linear Programming Duality Summer 2011 Optimization I Lecture 8 1 Duality recap Linear Programming Duality We motivated the dual of a linear program by thinking about the best possible lower bound on the optimal value we can achieve

More information

Statistical Approaches to Learning and Discovery

Statistical Approaches to Learning and Discovery Statistical Approaches to Learning and Discovery Graphical Models Zoubin Ghahramani & Teddy Seidenfeld zoubin@cs.cmu.edu & teddy@stat.cmu.edu CALD / CS / Statistics / Philosophy Carnegie Mellon University

More information

13 : Variational Inference: Loopy Belief Propagation and Mean Field

13 : 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 information

Study Notes on the Latent Dirichlet Allocation

Study 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 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

Regularization on Discrete Spaces

Regularization on Discrete Spaces Regularization on Discrete Spaces Dengyong Zhou and Bernhard Schölkopf Max Planck Institute for Biological Cybernetics Spemannstr. 38, 72076 Tuebingen, Germany {dengyong.zhou, bernhard.schoelkopf}@tuebingen.mpg.de

More information

Message Passing Algorithm with MAP Decoding on Zigzag Cycles for Non-binary LDPC Codes

Message Passing Algorithm with MAP Decoding on Zigzag Cycles for Non-binary LDPC Codes Message Passing Algorithm with MAP Decoding on Zigzag Cycles for Non-binary LDPC Codes Takayuki Nozaki 1, Kenta Kasai 2, Kohichi Sakaniwa 2 1 Kanagawa University 2 Tokyo Institute of Technology July 12th,

More information

A Simpler Max-Product Maximum Weight Matching Algorithm and the Auction Algorithm

A Simpler Max-Product Maximum Weight Matching Algorithm and the Auction Algorithm ISIT 006 Seattle USA July 9 14 006 A Simpler Max-Product Maximum Weight Matching Algorithm and the Auction Algorithm Mohsen Bayati Department of EE Stanford University Stanford CA 94305 Email: bayati@stanford.edu

More information

Corroborating Information from Disagreeing Views

Corroborating Information from Disagreeing Views Corroboration A. Galland WSDM 2010 1/26 Corroborating Information from Disagreeing Views Alban Galland 1 Serge Abiteboul 1 Amélie Marian 2 Pierre Senellart 3 1 INRIA Saclay Île-de-France 2 Rutgers University

More information

CHAPTER 6 : LITERATURE REVIEW

CHAPTER 6 : LITERATURE REVIEW CHAPTER 6 : LITERATURE REVIEW Chapter : LITERATURE REVIEW 77 M E A S U R I N G T H E E F F I C I E N C Y O F D E C I S I O N M A K I N G U N I T S A B S T R A C T A n o n l i n e a r ( n o n c o n v e

More information

P E R E N C O - C H R I S T M A S P A R T Y

P E R E N C O - C H R I S T M A S P A R T Y L E T T I C E L E T T I C E I S A F A M I L Y R U N C O M P A N Y S P A N N I N G T W O G E N E R A T I O N S A N D T H R E E D E C A D E S. B A S E D I N L O N D O N, W E H A V E T H E P E R F E C T R

More information

A Randomized Approach for Crowdsourcing in the Presence of Multiple Views

A Randomized Approach for Crowdsourcing in the Presence of Multiple Views A Randomized Approach for Crowdsourcing in the Presence of Multiple Views Presenter: Yao Zhou joint work with: Jingrui He - 1 - Roadmap Motivation Proposed framework: M2VW Experimental results Conclusion

More information

Notes on Markov Networks

Notes on Markov Networks Notes on Markov Networks Lili Mou moull12@sei.pku.edu.cn December, 2014 This note covers basic topics in Markov networks. We mainly talk about the formal definition, Gibbs sampling for inference, and maximum

More information

Communities Via Laplacian Matrices. Degree, Adjacency, and Laplacian Matrices Eigenvectors of Laplacian Matrices

Communities Via Laplacian Matrices. Degree, Adjacency, and Laplacian Matrices Eigenvectors of Laplacian Matrices Communities Via Laplacian Matrices Degree, Adjacency, and Laplacian Matrices Eigenvectors of Laplacian Matrices The Laplacian Approach As with betweenness approach, we want to divide a social graph into

More information

ECEN 689 Special Topics in Data Science for Communications Networks

ECEN 689 Special Topics in Data Science for Communications Networks ECEN 689 Special Topics in Data Science for Communications Networks Nick Duffield Department of Electrical & Computer Engineering Texas A&M University Lecture 8 Random Walks, Matrices and PageRank Graphs

More information

Part 1: Expectation Propagation

Part 1: Expectation Propagation Chalmers Machine Learning Summer School Approximate message passing and biomedicine Part 1: Expectation Propagation Tom Heskes Machine Learning Group, Institute for Computing and Information Sciences Radboud

More information

Expectation propagation as a way of life

Expectation propagation as a way of life Expectation propagation as a way of life Yingzhen Li Department of Engineering Feb. 2014 Yingzhen Li (Department of Engineering) Expectation propagation as a way of life Feb. 2014 1 / 9 Reference This

More information

Classification Semi-supervised learning based on network. Speakers: Hanwen Wang, Xinxin Huang, and Zeyu Li CS Winter

Classification Semi-supervised learning based on network. Speakers: Hanwen Wang, Xinxin Huang, and Zeyu Li CS Winter Classification Semi-supervised learning based on network Speakers: Hanwen Wang, Xinxin Huang, and Zeyu Li CS 249-2 2017 Winter Semi-Supervised Learning Using Gaussian Fields and Harmonic Functions Xiaojin

More information

Data Mining Techniques

Data Mining Techniques Data Mining Techniques CS 622 - Section 2 - Spring 27 Pre-final Review Jan-Willem van de Meent Feedback Feedback https://goo.gl/er7eo8 (also posted on Piazza) Also, please fill out your TRACE evaluations!

More information

Impossibility of Distributed Consensus with One Faulty Process

Impossibility of Distributed Consensus with One Faulty Process Impossibility of Distributed Consensus with One Faulty Process Journal of the ACM 32(2):374-382, April 1985. MJ Fischer, NA Lynch, MS Peterson. Won the 2002 Dijkstra Award (for influential paper in distributed

More information

Introduction to Low-Density Parity Check Codes. Brian Kurkoski

Introduction to Low-Density Parity Check Codes. Brian Kurkoski Introduction to Low-Density Parity Check Codes Brian Kurkoski kurkoski@ice.uec.ac.jp Outline: Low Density Parity Check Codes Review block codes History Low Density Parity Check Codes Gallager s LDPC code

More information

Lecture 3. 1 Polynomial-time algorithms for the maximum flow problem

Lecture 3. 1 Polynomial-time algorithms for the maximum flow problem ORIE 633 Network Flows August 30, 2007 Lecturer: David P. Williamson Lecture 3 Scribe: Gema Plaza-Martínez 1 Polynomial-time algorithms for the maximum flow problem 1.1 Introduction Let s turn now to considering

More information

ECE 592 Topics in Data Science

ECE 592 Topics in Data Science ECE 592 Topics in Data Science Final Fall 2017 December 11, 2017 Please remember to justify your answers carefully, and to staple your test sheet and answers together before submitting. Name: Student ID:

More information

Attentive Betweenness Centrality (ABC): Considering Options and Bandwidth when Measuring Criticality

Attentive Betweenness Centrality (ABC): Considering Options and Bandwidth when Measuring Criticality Attentive Betweenness Centrality (ABC): Considering Options and Bandwidth when Measuring Criticality Sibel Adalı, Xiaohui Lu, M. Magdon-Ismail September 5, 0 Who is the Most Critical? Would you use the

More information

Fault-Tolerant Consensus

Fault-Tolerant Consensus Fault-Tolerant Consensus CS556 - Panagiota Fatourou 1 Assumptions Consensus Denote by f the maximum number of processes that may fail. We call the system f-resilient Description of the Problem Each process

More information

Expectation Propagation Algorithm

Expectation 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 information

Information Geometric view of Belief Propagation

Information Geometric view of Belief Propagation Information Geometric view of Belief Propagation Yunshu Liu 2013-10-17 References: [1]. Shiro Ikeda, Toshiyuki Tanaka and Shun-ichi Amari, Stochastic reasoning, Free energy and Information Geometry, Neural

More information

Faloutsos, Tong ICDE, 2009

Faloutsos, Tong ICDE, 2009 Large Graph Mining: Patterns, Tools and Case Studies Christos Faloutsos Hanghang Tong CMU Copyright: Faloutsos, Tong (29) 2-1 Outline Part 1: Patterns Part 2: Matrix and Tensor Tools Part 3: Proximity

More information

Lecture 2. Judging the Performance of Classifiers. Nitin R. Patel

Lecture 2. Judging the Performance of Classifiers. Nitin R. Patel Lecture 2 Judging the Performance of Classifiers Nitin R. Patel 1 In this note we will examine the question of how to udge the usefulness of a classifier and how to compare different classifiers. Not only

More information

Illustration of the K2 Algorithm for Learning Bayes Net Structures

Illustration of the K2 Algorithm for Learning Bayes Net Structures Illustration of the K2 Algorithm for Learning Bayes Net Structures Prof. Carolina Ruiz Department of Computer Science, WPI ruiz@cs.wpi.edu http://www.cs.wpi.edu/ ruiz The purpose of this handout is to

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

abhi shelat

abhi shelat L15 4102.17.2016 abhi shelat Huffman image: wikimedia Alice m Bob m Alice m Bob MOSCOW President Vladimir V. Putin s typically theatrical order to withdraw the bulk of Russian forces from Syria, a process

More information

Recitation 9: Loopy BP

Recitation 9: Loopy BP Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science 6.438 Algorithms For Inference Fall 204 Recitation 9: Loopy BP General Comments. In terms of implementation,

More information

A Key Recovery Attack on MDPC with CCA Security Using Decoding Errors

A Key Recovery Attack on MDPC with CCA Security Using Decoding Errors A Key Recovery Attack on MDPC with CCA Security Using Decoding Errors Qian Guo Thomas Johansson Paul Stankovski Dept. of Electrical and Information Technology, Lund University ASIACRYPT 2016 Dec 8th, 2016

More information

Approximate Message Passing

Approximate Message Passing Approximate Message Passing Mohammad Emtiyaz Khan CS, UBC February 8, 2012 Abstract In this note, I summarize Sections 5.1 and 5.2 of Arian Maleki s PhD thesis. 1 Notation We denote scalars by small letters

More information

Multi-Dimensional Online Tracking

Multi-Dimensional Online Tracking Multi-Dimensional Online Tracking Ke Yi and Qin Zhang Hong Kong University of Science & Technology SODA 2009 January 4-6, 2009 1-1 A natural problem Bob: tracker f(t) g(t) Alice: observer (t, g(t)) t 2-1

More information

Linear Sketches A Useful Tool in Streaming and Compressive Sensing

Linear Sketches A Useful Tool in Streaming and Compressive Sensing Linear Sketches A Useful Tool in Streaming and Compressive Sensing Qin Zhang 1-1 Linear sketch Random linear projection M : R n R k that preserves properties of any v R n with high prob. where k n. M =

More information

7.5 Bipartite Matching

7.5 Bipartite Matching 7. Bipartite Matching Matching Matching. Input: undirected graph G = (V, E). M E is a matching if each node appears in at most edge in M. Max matching: find a max cardinality matching. Bipartite Matching

More information

Data Mining: Concepts and Techniques. (3 rd ed.) Chapter 6

Data Mining: Concepts and Techniques. (3 rd ed.) Chapter 6 Data Mining: Concepts and Techniques (3 rd ed.) Chapter 6 Jiawei Han, Micheline Kamber, and Jian Pei University of Illinois at Urbana-Champaign & Simon Fraser University 2013 Han, Kamber & Pei. All rights

More information

Bayesian Machine Learning - Lecture 7

Bayesian Machine Learning - Lecture 7 Bayesian Machine Learning - Lecture 7 Guido Sanguinetti Institute for Adaptive and Neural Computation School of Informatics University of Edinburgh gsanguin@inf.ed.ac.uk March 4, 2015 Today s lecture 1

More information

9. PSPACE 9. PSPACE. PSPACE complexity class quantified satisfiability planning problem PSPACE-complete

9. PSPACE 9. PSPACE. PSPACE complexity class quantified satisfiability planning problem PSPACE-complete Geography game Geography. Alice names capital city c of country she is in. Bob names a capital city c' that starts with the letter on which c ends. Alice and Bob repeat this game until one player is unable

More information

Recall: Matchings. Examples. K n,m, K n, Petersen graph, Q k ; graphs without perfect matching

Recall: Matchings. Examples. K n,m, K n, Petersen graph, Q k ; graphs without perfect matching Recall: Matchings A matching is a set of (non-loop) edges with no shared endpoints. The vertices incident to an edge of a matching M are saturated by M, the others are unsaturated. A perfect matching of

More information

An Optimal Algorithm for l 1 -Heavy Hitters in Insertion Streams and Related Problems

An Optimal Algorithm for l 1 -Heavy Hitters in Insertion Streams and Related Problems An Optimal Algorithm for l 1 -Heavy Hitters in Insertion Streams and Related Problems Arnab Bhattacharyya, Palash Dey, and David P. Woodruff Indian Institute of Science, Bangalore {arnabb,palash}@csa.iisc.ernet.in

More information

RaRE: Social Rank Regulated Large-scale Network Embedding

RaRE: Social Rank Regulated Large-scale Network Embedding RaRE: Social Rank Regulated Large-scale Network Embedding Authors: Yupeng Gu 1, Yizhou Sun 1, Yanen Li 2, Yang Yang 3 04/26/2018 The Web Conference, 2018 1 University of California, Los Angeles 2 Snapchat

More information

Web Structure Mining Nodes, Links and Influence

Web Structure Mining Nodes, Links and Influence Web Structure Mining Nodes, Links and Influence 1 Outline 1. Importance of nodes 1. Centrality 2. Prestige 3. Page Rank 4. Hubs and Authority 5. Metrics comparison 2. Link analysis 3. Influence model 1.

More information

9. PSPACE. PSPACE complexity class quantified satisfiability planning problem PSPACE-complete

9. PSPACE. PSPACE complexity class quantified satisfiability planning problem PSPACE-complete 9. PSPACE PSPACE complexity class quantified satisfiability planning problem PSPACE-complete Lecture slides by Kevin Wayne Copyright 2005 Pearson-Addison Wesley Copyright 2013 Kevin Wayne http://www.cs.princeton.edu/~wayne/kleinberg-tardos

More information

Convergence Analysis of the Variance in. Gaussian Belief Propagation

Convergence Analysis of the Variance in. Gaussian Belief Propagation 0.09/TSP.204.2345635, IEEE Transactions on Signal Processing Convergence Analysis of the Variance in Gaussian Belief Propagation Qinliang Su and Yik-Chung Wu Abstract It is known that Gaussian belief propagation

More information

The Generalized Distributive Law and Free Energy Minimization

The Generalized Distributive Law and Free Energy Minimization The Generalized Distributive Law and Free Energy Minimization Srinivas M. Aji Robert J. McEliece Rainfinity, Inc. Department of Electrical Engineering 87 N. Raymond Ave. Suite 200 California Institute

More information

Exact Solution of the Social Learning Model

Exact Solution of the Social Learning Model ICE2014, Shanghai Exact Solution of the Social Learning Model Jinshan Wu School of Systems Science, Beijing Normal University 06/01/2014 J. Wu (BNU) Exact Solution of Social Learning Shanghai 2014 1 /

More information

A Nearly Sublinear Approximation to exp{p}e i for Large Sparse Matrices from Social Networks

A Nearly Sublinear Approximation to exp{p}e i for Large Sparse Matrices from Social Networks A Nearly Sublinear Approximation to exp{p}e i for Large Sparse Matrices from Social Networks Kyle Kloster and David F. Gleich Purdue University December 14, 2013 Supported by NSF CAREER 1149756-CCF Kyle

More information

Reconstruction in the Sparse Labeled Stochastic Block Model

Reconstruction in the Sparse Labeled Stochastic Block Model Reconstruction in the Sparse Labeled Stochastic Block Model Marc Lelarge 1 Laurent Massoulié 2 Jiaming Xu 3 1 INRIA-ENS 2 INRIA-Microsoft Research Joint Centre 3 University of Illinois, Urbana-Champaign

More information

Mining of Massive Datasets Jure Leskovec, AnandRajaraman, Jeff Ullman Stanford University

Mining of Massive Datasets Jure Leskovec, AnandRajaraman, Jeff Ullman Stanford University Note to other teachers and users of these slides: We would be delighted if you found this our material useful in giving your own lectures. Feel free to use these slides verbatim, or to modify them to fit

More information

Final Examination CS 540-2: Introduction to Artificial Intelligence

Final Examination CS 540-2: Introduction to Artificial Intelligence Final Examination CS 540-2: Introduction to Artificial Intelligence May 7, 2017 LAST NAME: SOLUTIONS FIRST NAME: Problem Score Max Score 1 14 2 10 3 6 4 10 5 11 6 9 7 8 9 10 8 12 12 8 Total 100 1 of 11

More information

Data Mining: Concepts and Techniques. (3 rd ed.) Chapter 8. Chapter 8. Classification: Basic Concepts

Data Mining: Concepts and Techniques. (3 rd ed.) Chapter 8. Chapter 8. Classification: Basic Concepts Data Mining: Concepts and Techniques (3 rd ed.) Chapter 8 1 Chapter 8. Classification: Basic Concepts Classification: Basic Concepts Decision Tree Induction Bayes Classification Methods Rule-Based Classification

More information

University of Chicago Autumn 2003 CS Markov Chain Monte Carlo Methods. Lecture 7: November 11, 2003 Estimating the permanent Eric Vigoda

University of Chicago Autumn 2003 CS Markov Chain Monte Carlo Methods. Lecture 7: November 11, 2003 Estimating the permanent Eric Vigoda University of Chicago Autumn 2003 CS37101-1 Markov Chain Monte Carlo Methods Lecture 7: November 11, 2003 Estimating the permanent Eric Vigoda We refer the reader to Jerrum s book [1] for the analysis

More information

Enabling Accurate Analysis of Private Network Data

Enabling Accurate Analysis of Private Network Data Enabling Accurate Analysis of Private Network Data Michael Hay Joint work with Gerome Miklau, David Jensen, Chao Li, Don Towsley University of Massachusetts, Amherst Vibhor Rastogi, Dan Suciu University

More information

Probabilistic Graphical Models

Probabilistic 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 information

PathSelClus: Integrating Meta-Path Selection with User-Guided Object Clustering in Heterogeneous Information Networks

PathSelClus: Integrating Meta-Path Selection with User-Guided Object Clustering in Heterogeneous Information Networks PathSelClus: Integrating Meta-Path Selection with User-Guided Object Clustering in Heterogeneous Information Networks YIZHOU SUN, BRANDON NORICK, and JIAWEI HAN, University of Illinois at Urbana-Champaign

More information

CS Lecture 19. Exponential Families & Expectation Propagation

CS Lecture 19. Exponential Families & Expectation Propagation CS 6347 Lecture 19 Exponential Families & Expectation Propagation Discrete State Spaces We have been focusing on the case of MRFs over discrete state spaces Probability distributions over discrete spaces

More information

MobiHoc 2014 MINIMUM-SIZED INFLUENTIAL NODE SET SELECTION FOR SOCIAL NETWORKS UNDER THE INDEPENDENT CASCADE MODEL

MobiHoc 2014 MINIMUM-SIZED INFLUENTIAL NODE SET SELECTION FOR SOCIAL NETWORKS UNDER THE INDEPENDENT CASCADE MODEL MobiHoc 2014 MINIMUM-SIZED INFLUENTIAL NODE SET SELECTION FOR SOCIAL NETWORKS UNDER THE INDEPENDENT CASCADE MODEL Jing (Selena) He Department of Computer Science, Kennesaw State University Shouling Ji,

More information

Random Target Lemma. Sucharita Jayanti. February 21, 2013

Random Target Lemma. Sucharita Jayanti. February 21, 2013 Random Target Lemma Sucharita Jayanti February 21, 2013 Abstract The well known Random Target Lemma states that if one starts at a node i in a connected graph, selects a target node at random (according

More information

CMU SCS. Large Graph Mining. Christos Faloutsos CMU

CMU SCS. Large Graph Mining. Christos Faloutsos CMU Large Graph Mining Christos Faloutsos CMU Thank you! Hillol Kargupta NGDM 2007 C. Faloutsos 2 Outline Problem definition / Motivation Static & dynamic laws; generators Tools: CenterPiece graphs; fraud

More information

Solve the following equations. Show all work to receive credit. No decimal answers. 8) 4x 2 = 100

Solve the following equations. Show all work to receive credit. No decimal answers. 8) 4x 2 = 100 Algebra 2 1.1 Worksheet Name Solve the following equations. Show all work to receive credit. No decimal answers. 1) 3x 5(2 4x) = 18 2) 17 + 11x = -19x 25 3) 2 6x+9 b 4 = 7 4) = 2x 3 4 5) 3 = 5 7 x x+1

More information

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials by Phillip Krahenbuhl and Vladlen Koltun Presented by Adam Stambler Multi-class image segmentation Assign a class label to each

More information

Computer Vision Group Prof. Daniel Cremers. 10a. Markov Chain Monte Carlo

Computer Vision Group Prof. Daniel Cremers. 10a. Markov Chain Monte Carlo Group Prof. Daniel Cremers 10a. Markov Chain Monte Carlo Markov Chain Monte Carlo In high-dimensional spaces, rejection sampling and importance sampling are very inefficient An alternative is Markov Chain

More information

Unifying Guilt-by-Association Approaches: Theorems and Fast Algorithms

Unifying Guilt-by-Association Approaches: Theorems and Fast Algorithms Unifying Guilt-by-Association Approaches: Theorems and Fast Algorithms Danai Koutra 1, Tai-You Ke 2,U.Kang 1, Duen Horng (Polo) Chau 1, Hsing-Kuo Kenneth Pao 2, and Christos Faloutsos 1 1 School of Computer

More information

CS 580: Algorithm Design and Analysis. Jeremiah Blocki Purdue University Spring 2018

CS 580: Algorithm Design and Analysis. Jeremiah Blocki Purdue University Spring 2018 CS 580: Algorithm Design and Analysis Jeremiah Blocki Purdue University Spring 2018 Chapter 9 PSPACE: A Class of Problems Beyond NP Slides by Kevin Wayne. Copyright @ 2005 Pearson-Addison Wesley. All rights

More information

Readings: K&F: 16.3, 16.4, Graphical Models Carlos Guestrin Carnegie Mellon University October 6 th, 2008

Readings: 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 information

UNDERSTANDING BELIEF PROPOGATION AND ITS GENERALIZATIONS

UNDERSTANDING BELIEF PROPOGATION AND ITS GENERALIZATIONS UNDERSTANDING BELIEF PROPOGATION AND ITS GENERALIZATIONS JONATHAN YEDIDIA, WILLIAM FREEMAN, YAIR WEISS 2001 MERL TECH REPORT Kristin Branson and Ian Fasel June 11, 2003 1. Inference Inference problems

More information

Spectral Graph Theory Lecture 2. The Laplacian. Daniel A. Spielman September 4, x T M x. ψ i = arg min

Spectral Graph Theory Lecture 2. The Laplacian. Daniel A. Spielman September 4, x T M x. ψ i = arg min Spectral Graph Theory Lecture 2 The Laplacian Daniel A. Spielman September 4, 2015 Disclaimer These notes are not necessarily an accurate representation of what happened in class. The notes written before

More information

Exponential Families

Exponential 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 information

Privacy and Fault-Tolerance in Distributed Optimization. Nitin Vaidya University of Illinois at Urbana-Champaign

Privacy and Fault-Tolerance in Distributed Optimization. Nitin Vaidya University of Illinois at Urbana-Champaign Privacy and Fault-Tolerance in Distributed Optimization Nitin Vaidya University of Illinois at Urbana-Champaign Acknowledgements Shripad Gade Lili Su argmin x2x SX i=1 i f i (x) Applications g f i (x)

More information

Part V. Matchings. Matching. 19 Augmenting Paths for Matchings. 18 Bipartite Matching via Flows

Part V. Matchings. Matching. 19 Augmenting Paths for Matchings. 18 Bipartite Matching via Flows Matching Input: undirected graph G = (V, E). M E is a matching if each node appears in at most one Part V edge in M. Maximum Matching: find a matching of maximum cardinality Matchings Ernst Mayr, Harald

More information

Adaptive Cut Generation for Improved Linear Programming Decoding of Binary Linear Codes

Adaptive Cut Generation for Improved Linear Programming Decoding of Binary Linear Codes Adaptive Cut Generation for Improved Linear Programming Decoding of Binary Linear Codes Xiaojie Zhang and Paul H. Siegel University of California, San Diego, La Jolla, CA 9093, U Email:{ericzhang, psiegel}@ucsd.edu

More information

Overlapping Communities

Overlapping Communities Overlapping Communities Davide Mottin HassoPlattner Institute Graph Mining course Winter Semester 2017 Acknowledgements Most of this lecture is taken from: http://web.stanford.edu/class/cs224w/slides GRAPH

More information

CMPUT651: Differential Privacy

CMPUT651: Differential Privacy CMPUT65: Differential Privacy Homework assignment # 2 Due date: Apr. 3rd, 208 Discussion and the exchange of ideas are essential to doing academic work. For assignments in this course, you are encouraged

More information