Exact Inference. Factor Graphs through Max-Sum Algorithm Figures from Bishop PRML Sec. 8.3/8.4. x 3. f s. x 2. x 1
|
|
- Lionel Ray
- 5 years ago
- Views:
Transcription
1 Exact Inference x 1 x 3 x 2 f s Geoffrey Roeder roeder@cs.toronto.edu 8 February 2018 Factor Graphs through Max-Sum Algorithm Figures from Bishop PRML Sec. 8.3/8.4
2 Building Blocks UGMs, Cliques, Factor Graphs
3 Markov Random Fields / UGMs Parameterization: maximal cliques x 1 x 2 x 3 x 4
4 Example: Equivalent DGM and UGM x 1 x 2 x N 1 x N x 1 x 2 x N x N 1 p(x) =p(x 1 )p(x 2 x 1 )p(x 3 x 2 ) p(x N x N 1 ). vert this to an undirected graph representation, p(x) = 1 Z ψ 1,2(x 1,x 2 )ψ 2,3 (x 2,x 3 ) ψ N 1,N (x N 1,x N ) ψ 1,2 (x 1,x 2 ) = p(x 1 )p(x 2 x 1 ) ψ 2,3 (x 2,x 3 ) = p(x 3 x 2 ). ψ N 1,N (x N 1,x N ) = p(x N x N 1 )
5 Conversion: Moralization (Marry the Parents of Every Child) x 1 x 3 x 1 x 3 x 2 x 2 (a) x 4 (b) x 4 p(x) =p(x 1 )p(x 2 )p(x 3 )p(x 4 x 1,x 2,x 3 ).
6 DGMs and UGMs represent distinct distributions A B C F A B C D U P D
7 Motivations: Exact Inference in a Chain
8 Query probability of a configuration for node X_n: p(x_n) x 1 x 2 x N 1 x N x 1 x 2 x N x N 1 p(x) = 1 Z ψ 1,2(x 1,x 2 )ψ 2,3 (x 2,x 3 ) ψ N 1,N (x N 1,x N ). p(x n )= x 1 x n 1 p(x). x n+1 x N
9 Query probability of a configuration for node X_n: p(x_n) x 1 x 2 x N 1 x N x 1 x 2 x N x N 1 p(x) = 1 Z ψ 1,2(x 1,x 2 )ψ 2,3 (x 2,x 3 ) ψ N 1,N (x N 1,x N ). p(x n )= x 1 x n 1 p(x). x n+1 x N Naively: N variables, K states per variable: computation complexity? p(x n )= x 1 x n 1 p(x). x n+1 x N
10 Query probability of a configuration for node X_n: p(x_n) x 1 x 2 x N 1 x N x 1 x 2 x N x N 1 p(x) = 1 Z ψ 1,2(x 1,x 2 )ψ 2,3 (x 2,x 3 ) ψ N 1,N (x N 1,x N ). p(x n )= x 1 x n 1 p(x). x n+1 x N We ignored the conditional independence! Notice for x_n: summation ψ N 1,N (x N 1,x N ) x N
11 Be clever about order of computation: x 1 x 2 x N 1 x N p(x n )= 1 Z [ [ ]] ψ n 1,n (x n 1,x n ) ψ 2,3 (x 2,x 3 ) ψ 1,2 (x 1,x 2 ) xn 1 x 2 x 1 }{{} µ α (x n ) [ ] ψ n,n+1 (x n,x n+1 ) ψ N 1,N (x N 1,x N ). (8.52) xn+1 x N }{{} µ β (x n ) p(x n )= 1 Z µ α(x n )µ β (x n )
12 Be clever about order of computation: x 1 x 2 x N 1 x N µ α (x n 1 ) µ α (x n ) µ β (x n ) µ β (x n+1 ) x 1 x n 1 x n x n+1 x N p(x n )= 1 Z µ α(x n )µ β (x n )
13 We get joint marginals over variables, too: x 1 x 2 x N 1 x N µ α (x n 1 ) µ α (x n ) µ β (x n ) µ β (x n+1 ) x 1 x n 1 x n x n+1 x N p(x n 1,x n )= 1 Z µ α(x n 1 )ψ n 1,n (x n 1,x n )µ β (x n ). obtain the joint distributions over all of the sets of var
14 Factor Graph Review
15 x 1 x 2 x 3 f a f b f c f d p(x) = Y s f(x s ) p(x) =f a (x 1,x 2 )f b (x 1,x 2 )f c (x 2,x 3 )f d (x 3 )
16 x 1 x 2 x 1 x 2 f x 1 x 2 f c f a f b (a) x 3 (b) x 3 (c) x 3 (a) p(x 1 )p(x 2 )p(x 3 x 1,x 2 ) (b) f(x 1,x 2,x 3 )=p(x 1 )p(x 2 )p(x 3 x 1,x 2 ) (c) f a (x 1 )=p(x 1 ),f b (x 2 )=p(x 2 ),f c (x 1,x 2,x 3 )=p(x 3 x 1,x 2 )
17 x 1 x 2 x 1 x 2 x 1 x 2 f a f(x 1,x 2,x 3 ) f b f c (a) x 3 (b) x 3 (c) x 3
18 Sum-Product Algorithm Generalize Exact Inference in Chains to Tree-Structured PGMs
19 Problem setup: notation Fs(x, Xs) f s µ fs x(x) x p(x) = x\x p(x) = p(x) F s (x, X s ) s ne(x) ne(x): set of factor nodes that are neighbours of x X_s: set of all variables in the subtree connected to the variable node x via factor node f_s F_s(x, X_s): the product of all the factors in the group associated with factor f_s
20 Problem setup: notation Fs(x, Xs) f s µ fs x(x) x p(x) = = [ s ne(x) s ne(x) F s (x, X s ) X s µ fs x(x). ] ne(x): set of factor nodes that are neighbours of x X_s: set of all variables in the subtree connected to the variable node x via factor node f_s F_s(x, X_s): the product of all the factors in the group associated with factor f_s
21 Problem setup: notation Fs(x, Xs) f s µ fs x(x) x p(x) = = [ s ne(x) s ne(x) F s (x, X s ) X s µ fs x(x). ] We evaluate the marginal p(x) as product of messages from surrounding factors!
22 Factor messages: decomposition [ ] p(x) = F s (x, X s ) s ne(x) X s = µ fs x(x). s ne(x) Each factor is itself described by a factor sub-graph, so we can decompose: F s (x, X s )=f s (x, x 1,...,x M )G 1 (x 1,X s1 )...G M (x M,X sm ) (Each variable associated with f_x is {x, x_1,, x_m}) Rewriting the factor-to-variable message: [ ] µ fs x(x) =... f s (x, x 1,...,x M ) G m (x m,x sm ) x 1 x M X xm = x 1... x M f s (x, x 1,...,x M ) m ne(f s )\x m ne(f s )\x µ xm f s (x m ) (8.66)
23 Factor messages: decomposition [ ] p(x) = F s (x, X s ) s ne(x) X s = µ fs x(x). s ne(x) Each factor is itself described by a factor sub-graph, so we can decompose: F s (x, X s )=f s (x, x 1,...,x M )G 1 (x 1,X s1 )...G M (x M,X sm ) (Each variable associated with f_x is {x, x_1,, x_m}) Rewriting the factor-to-variable message: [ ] µ fs x(x) =... f s (x, x 1,...,x M ) G m (x m,x sm ) x 1 x M X xm = x 1... x M f s (x, x 1,...,x M ) m ne(f s )\x m ne(f s )\x µ xm f s (x m ) (8.66)
24 Factor-to-variable messages: decomposition [ ] p(x) = F s (x, X s ) s ne(x) X s = µ fs x(x). s ne(x) Each factor is itself described by a factor sub-graph, so we can decompose: F s (x, X s )=f s (x, x 1,...,x M )G 1 (x 1,X s1 )...G M (x M,X sm ) (Each variable associated with f_x is {x, x_1,, x_m}) Rewriting the factor-to-variable message: [ ] µ fs x(x) =... f s (x, x 1,...,x M ) G m (x m,x sm ) x 1 x M X xm = x 1... x M f s (x, x 1,...,x M ) m ne(f s )\x m ne(f s )\x µ xm f s (x m ) (8.66) µ xm f s (x m ) X sm G m (x m,x sm ).
25 Variable-to-factor messages: decomposition x M µ xm f s (x M ) f s µ fs x(x) x x m G m (x m,x sm ) µ xm f s (x m ) X sm G m (x m,x sm ).
26 Factor-to-variable messages: one step back towards the leaves f L x m f s f l F l (x m,x ml ) F s (x, X s )=f s (x, x 1,...,x M )G 1 (x 1,X s1 )...G M (x M,X sm ) G m (x m,x sm )= ] F l (x m,x ml ) l ne(x m )\f s
27 Factor-to-variable messages: one step back towards the leaves f L x m f s f l F l (x m,x ml ) µ xm f s (x m ) = = l ne(x m )\f s ] [ X ml F l (x m,x ml ) l ne(x m )\f s µ fl x m (x m ) ]
28 Sum-Product Initialization at Leaves µ x f (x) =1 µ f x (x) =f(x) x f f x Sum-Product: Marginal distribution over x p(x) = = [ s ne(x) s ne(x) F s (x, X s ) X s µ fs x(x). ] See Bishop p. 409 for a fully worked, simple example!
Inference in Bayesian Networks
Andrea Passerini passerini@disi.unitn.it Machine Learning Inference in graphical models Description Assume we have evidence e on the state of a subset of variables E in the model (i.e. Bayesian Network)
More informationFactor Graphs. Seungjin Choi
Factor Graphs Seungjin Choi Department of Computer Science and Engineering Pohang University of Science and Technology 77 Cheongam-ro, Nam-gu, Pohang 37673, Korea seungjin@postech.ac.kr 1 / 17 Tanner Graphs
More informationSum-Product: Message Passing Belief Propagation
Sum-Product: Message Passing Belief Propagation Probabilistic Graphical Models Sharif University of Technology Spring 2017 Soleymani All single-node marginals If we need the full set of marginals, repeating
More informationSum-Product: Message Passing Belief Propagation
Sum-Product: Message Passing Belief Propagation 40-956 Advanced Topics in AI: Probabilistic Graphical Models Sharif University of Technology Soleymani Spring 2015 All single-node marginals If we need the
More informationProbabilistic Graphical Models
CS420, Machine Learning, Lecture 8 Probabilistic Graphical Models Weinan Zhang Shanghai Jiao Tong University http://wnzhang.net http://wnzhang.net/teaching/cs420/index.html Content of This Lecture Introduction
More informationCOMP90051 Statistical Machine Learning
COMP90051 Statistical Machine Learning Semester 2, 2017 Lecturer: Trevor Cohn 22. PGM Probabilistic Inference Probabilistic inference on PGMs Computing marginal and conditional distributions from the joint
More informationGibbs Fields: Inference and Relation to Bayes Networks
Statistical Techniques in Robotics (16-831, F10) Lecture#08 (Thursday September 16) Gibbs Fields: Inference and Relation to ayes Networks Lecturer: rew agnell Scribe:ebadeepta ey 1 1 Inference on Gibbs
More informationGlobal Joint Distribution Factorizes into Local Marginal Distributions on Tree-Structured Graphs
Teaching Note October 26, 2007 Global Joint Distribution Factorizes into Local Marginal Distributions on Tree-Structured Graphs Xinhua Zhang Xinhua.Zhang@anu.edu.au Research School of Information Sciences
More informationUGM Crash Course: Conditional Inference and Cutset Conditioning
UGM Crash Course: Conditional Inference and Cutset Conditioning Julie Nutini August 19 th, 2015 1 / 25 Conditional UGM 2 / 25 We know the value of one or more random variables i.e., we have observations,
More informationExact Inference (9/30/13) 2 A brief review of Forward-Backward and EM for HMMs
STA561: Probabilistic machine learning Exact Inference (9/30/13) Lecturer: Barbara Engelhardt Scribes: Jiawei Liang, He Jiang, Brittany Cohen 1 Validation for Clustering If we have two centroids, η 1 and
More informationModelli Grafici Probabilistici (2): concetti generali
Modelli Grafici Probabilistici (2): concetti generali Corso di Modelli di Computazione Affettiva Prof. Giuseppe Boccignone Dipartimento di Informatica Università di Milano boccignone@di.unimi.it Giuseppe.Boccignone@unimi.it
More informationMachine Learning
Machine Learning 10-601 Tom M. Mitchell Machine Learning Department Carnegie Mellon University October 13, 2011 Today: Graphical models Bayes Nets: Conditional independencies Inference Learning Readings:
More informationA start of Variational Methods for ERGM Ranran Wang, UW
A start of Variational Methods for ERGM Ranran Wang, UW MURI-UCI April 24, 2009 Outline A start of Variational Methods for ERGM [1] Introduction to ERGM Current methods of parameter estimation: MCMCMLE:
More informationComputer Vision Group Prof. Daniel Cremers. 7. Sequential Data
Group Prof. Daniel Cremers 7. Sequential Data Bayes Filter (Rep.) We can describe the overall process using a Dynamic Bayes Network: This incorporates the following Markov assumptions: (measurement) (state)!2
More informationMachine Learning. Graphical Models. Marc Toussaint University of Stuttgart Summer 2015
Machine Learning Graphical Models Marc Toussaint University of Stuttgart Summer 2015 Outline A. Introduction Motivation and definition of Bayes Nets Conditional independence in Bayes Nets Examples B. Inference
More informationMore Advanced Single Machine Models. University at Buffalo IE661 Scheduling Theory 1
More Advanced Single Machine Models University at Buffalo IE661 Scheduling Theory 1 Total Earliness And Tardiness Non-regular performance measures Ej + Tj Early jobs (Set j 1 ) and Late jobs (Set j 2 )
More informationA potentially useful approach to model nonlinearities in time series is to assume different behavior (structural break) in different subsamples
1.3 Regime switching models A potentially useful approach to model nonlinearities in time series is to assume different behavior (structural break) in different subsamples (or regimes). If the dates, the
More informationSTAT/MATH 395 PROBABILITY II
STAT/MATH 395 PROBABILITY II Distribution of Random Samples & Limit Theorems Néhémy Lim University of Washington Winter 2017 Outline Distribution of i.i.d. Samples Convergence of random variables The Laws
More informationDefinition 4.1. In a stochastic process T is called a stopping time if you can tell when it happens.
102 OPTIMAL STOPPING TIME 4. Optimal Stopping Time 4.1. Definitions. On the first day I explained the basic problem using one example in the book. On the second day I explained how the solution to the
More information- 1 - **** d(lns) = (µ (1/2)σ 2 )dt + σdw t
- 1 - **** These answers indicate the solutions to the 2014 exam questions. Obviously you should plot graphs where I have simply described the key features. It is important when plotting graphs to label
More informationVersion A. Problem 1. Let X be the continuous random variable defined by the following pdf: 1 x/2 when 0 x 2, f(x) = 0 otherwise.
Math 224 Q Exam 3A Fall 217 Tues Dec 12 Version A Problem 1. Let X be the continuous random variable defined by the following pdf: { 1 x/2 when x 2, f(x) otherwise. (a) Compute the mean µ E[X]. E[X] x
More informationPortfolio Risk Management and Linear Factor Models
Chapter 9 Portfolio Risk Management and Linear Factor Models 9.1 Portfolio Risk Measures There are many quantities introduced over the years to measure the level of risk that a portfolio carries, and each
More informationBayes s Rule Example. defective. An MP3 player is selected at random and found to be defective. What is the probability it came from Factory I?
Bayes s Rule Example A company manufactures MP3 players at two factories. Factory I produces 60% of the MP3 players and Factory II produces 40%. Two percent of the MP3 players produced at Factory I are
More informationCS 294-2, Grouping and Recognition (Prof. Jitendra Malik) Aug 30, 1999 Lecture #3 (Maximum likelihood framework) DRAFT Notes by Joshua Levy ffl Maximu
CS 294-2, Grouping and Recognition (Prof. Jitendra Malik) Aug 30, 1999 Lecture #3 (Maximum likelihood framework) DRAFT Notes by Joshua Levy l Maximum likelihood framework The estimation problem Maximum
More informationCPSC 540: Machine Learning
CPSC 540: Machine Learning Monte Carlo Methods Mark Schmidt University of British Columbia Winter 2018 Last Time: Markov Chains We can use Markov chains for density estimation, p(x) = p(x 1 ) }{{} d p(x
More informationStratified Sampling in Monte Carlo Simulation: Motivation, Design, and Sampling Error
South Texas Project Risk- Informed GSI- 191 Evaluation Stratified Sampling in Monte Carlo Simulation: Motivation, Design, and Sampling Error Document: STP- RIGSI191- ARAI.03 Revision: 1 Date: September
More informationIs Greedy Coordinate Descent a Terrible Algorithm?
Is Greedy Coordinate Descent a Terrible Algorithm? Julie Nutini, Mark Schmidt, Issam Laradji, Michael Friedlander, Hoyt Koepke University of British Columbia Optimization and Big Data, 2015 Context: Random
More informationCPSC 540: Machine Learning
CPSC 540: Machine Learning Monte Carlo Methods Mark Schmidt University of British Columbia Winter 2019 Last Time: Markov Chains We can use Markov chains for density estimation, d p(x) = p(x 1 ) p(x }{{}
More informationMachine Learning in Computer Vision Markov Random Fields Part II
Machine Learning in Computer Vision Markov Random Fields Part II Oren Freifeld Computer Science, Ben-Gurion University March 22, 2018 Mar 22, 2018 1 / 40 1 Some MRF Computations 2 Mar 22, 2018 2 / 40 Few
More information15 : Approximate Inference: Monte Carlo Methods
10-708: Probabilistic Graphical Models 10-708, Spring 2016 15 : Approximate Inference: Monte Carlo Methods Lecturer: Eric P. Xing Scribes: Binxuan Huang, Yotam Hechtlinger, Fuchen Liu 1 Introduction to
More informationLecture Notes 6. Assume F belongs to a family of distributions, (e.g. F is Normal), indexed by some parameter θ.
Sufficient Statistics Lecture Notes 6 Sufficiency Data reduction in terms of a particular statistic can be thought of as a partition of the sample space X. Definition T is sufficient for θ if the conditional
More informationExam M Fall 2005 PRELIMINARY ANSWER KEY
Exam M Fall 005 PRELIMINARY ANSWER KEY Question # Answer Question # Answer 1 C 1 E C B 3 C 3 E 4 D 4 E 5 C 5 C 6 B 6 E 7 A 7 E 8 D 8 D 9 B 9 A 10 A 30 D 11 A 31 A 1 A 3 A 13 D 33 B 14 C 34 C 15 A 35 A
More informationDynamic Programming (DP) Massimo Paolucci University of Genova
Dynamic Programming (DP) Massimo Paolucci University of Genova DP cannot be applied to each kind of problem In particular, it is a solution method for problems defined over stages For each stage a subproblem
More informationThe exam is closed book, closed calculator, and closed notes except your three crib sheets.
CS 188 Spring 2016 Introduction to Artificial Intelligence Final V2 You have approximately 2 hours and 50 minutes. The exam is closed book, closed calculator, and closed notes except your three crib sheets.
More informationCSE 312 Winter Learning From Data: Maximum Likelihood Estimators (MLE)
CSE 312 Winter 2017 Learning From Data: Maximum Likelihood Estimators (MLE) 1 Parameter Estimation Given: independent samples x1, x2,..., xn from a parametric distribution f(x θ) Goal: estimate θ. Not
More informationDirect Methods for linear systems Ax = b basic point: easy to solve triangular systems
NLA p.1/13 Direct Methods for linear systems Ax = b basic point: easy to solve triangular systems... 0 0 0 etc. a n 1,n 1 x n 1 = b n 1 a n 1,n x n solve a n,n x n = b n then back substitution: takes n
More informationThe exam is closed book, closed calculator, and closed notes except your one-page crib sheet.
CS 188 Summer 2015 Introduction to Artificial Intelligence Midterm 2 You have approximately 80 minutes. The exam is closed book, closed calculator, and closed notes except your one-page crib sheet. Mark
More informationRandom Tree Method. Monte Carlo Methods in Financial Engineering
Random Tree Method Monte Carlo Methods in Financial Engineering What is it for? solve full optimal stopping problem & estimate value of the American option simulate paths of underlying Markov chain produces
More information2 all subsequent nodes. 252 all subsequent nodes. 401 all subsequent nodes. 398 all subsequent nodes. 330 all subsequent nodes
¼ À ÈÌ Ê ½¾ ÈÊÇ Ä ÅË ½µ ½¾º¾¹½ ¾µ ½¾º¾¹ µ ½¾º¾¹ µ ½¾º¾¹ µ ½¾º ¹ µ ½¾º ¹ µ ½¾º ¹¾ µ ½¾º ¹ µ ½¾¹¾ ½¼µ ½¾¹ ½ (1) CLR 12.2-1 Based on the structure of the binary tree, and the procedure of Tree-Search, any
More informationIntroduction to Functions Section 2.1
Introduction to Functions Section 2.1 Notation Evaluation Solving Unit of measurement 1 Introductory Example: Fill the gas tank Your gas tank holds 12 gallons, but right now you re running on empty. As
More informationCS 174: Combinatorics and Discrete Probability Fall Homework 5. Due: Thursday, October 4, 2012 by 9:30am
CS 74: Combinatorics and Discrete Probability Fall 0 Homework 5 Due: Thursday, October 4, 0 by 9:30am Instructions: You should upload your homework solutions on bspace. You are strongly encouraged to type
More informationLecture 1: Lucas Model and Asset Pricing
Lecture 1: Lucas Model and Asset Pricing Economics 714, Spring 2018 1 Asset Pricing 1.1 Lucas (1978) Asset Pricing Model We assume that there are a large number of identical agents, modeled as a representative
More informationSequential Decision Making
Sequential Decision Making Dynamic programming Christos Dimitrakakis Intelligent Autonomous Systems, IvI, University of Amsterdam, The Netherlands March 18, 2008 Introduction Some examples Dynamic programming
More informationEC316a: Advanced Scientific Computation, Fall Discrete time, continuous state dynamic models: solution methods
EC316a: Advanced Scientific Computation, Fall 2003 Notes Section 4 Discrete time, continuous state dynamic models: solution methods We consider now solution methods for discrete time models in which decisions
More informationTolerance Intervals for Any Data (Nonparametric)
Chapter 831 Tolerance Intervals for Any Data (Nonparametric) Introduction This routine calculates the sample size needed to obtain a specified coverage of a β-content tolerance interval at a stated confidence
More informationClass 12. Daniel B. Rowe, Ph.D. Department of Mathematics, Statistics, and Computer Science. Marquette University MATH 1700
Class 12 Daniel B. Rowe, Ph.D. Department of Mathematics, Statistics, and Computer Science Copyright 2017 by D.B. Rowe 1 Agenda: Recap Chapter 6.1-6.2 Lecture Chapter 6.3-6.5 Problem Solving Session. 2
More informationA relation on 132-avoiding permutation patterns
Discrete Mathematics and Theoretical Computer Science DMTCS vol. VOL, 205, 285 302 A relation on 32-avoiding permutation patterns Natalie Aisbett School of Mathematics and Statistics, University of Sydney,
More informationThe Uniform Distribution
The Uniform Distribution EXAMPLE 1 The previous problem is an example of the uniform probability distribution. Illustrate the uniform distribution. The data that follows are 55 smiling times, in seconds,
More informationThe Normal Distribution. (Ch 4.3)
5 The Normal Distribution (Ch 4.3) The Normal Distribution The normal distribution is probably the most important distribution in all of probability and statistics. Many populations have distributions
More informationBrownian Motion, the Gaussian Lévy Process
Brownian Motion, the Gaussian Lévy Process Deconstructing Brownian Motion: My construction of Brownian motion is based on an idea of Lévy s; and in order to exlain Lévy s idea, I will begin with the following
More informationCS221 / Spring 2018 / Sadigh. Lecture 9: Games I
CS221 / Spring 2018 / Sadigh Lecture 9: Games I Course plan Search problems Markov decision processes Adversarial games Constraint satisfaction problems Bayesian networks Reflex States Variables Logic
More informationAsymptotic methods in risk management. Advances in Financial Mathematics
Asymptotic methods in risk management Peter Tankov Based on joint work with A. Gulisashvili Advances in Financial Mathematics Paris, January 7 10, 2014 Peter Tankov (Université Paris Diderot) Asymptotic
More informationImportance Sampling. Sargur N. Srihari
Importance Sampling Sargur N. srihari@cedar.buffalo.edu 1 Topics in Monte Carlo Methods 1. Sampling and Monte Carlo Methods 2. Importance Sampling 3. Markov Chain Monte Carlo Methods 4. Gibbs Sampling
More informationDynamic Portfolio Execution Detailed Proofs
Dynamic Portfolio Execution Detailed Proofs Gerry Tsoukalas, Jiang Wang, Kay Giesecke March 16, 2014 1 Proofs Lemma 1 (Temporary Price Impact) A buy order of size x being executed against i s ask-side
More informationLecture 9: Games I. Course plan. A simple game. Roadmap. Machine learning. Example: game 1
Lecture 9: Games I Course plan Search problems Markov decision processes Adversarial games Constraint satisfaction problems Bayesian networks Reflex States Variables Logic Low-level intelligence Machine
More informationParticle methods and the pricing of American options
Particle methods and the pricing of American options Peng HU Oxford-Man Institute April 29, 2013 Joint works with P. Del Moral, N. Oudjane & B. Rémillard P. HU (OMI) University of Oxford 1 / 46 Outline
More informationAnother Variant of 3sat. 3sat. 3sat Is NP-Complete. The Proof (concluded)
3sat k-sat, where k Z +, is the special case of sat. The formula is in CNF and all clauses have exactly k literals (repetition of literals is allowed). For example, (x 1 x 2 x 3 ) (x 1 x 1 x 2 ) (x 1 x
More informationHandout 8: Introduction to Stochastic Dynamic Programming. 2 Examples of Stochastic Dynamic Programming Problems
SEEM 3470: Dynamic Optimization and Applications 2013 14 Second Term Handout 8: Introduction to Stochastic Dynamic Programming Instructor: Shiqian Ma March 10, 2014 Suggested Reading: Chapter 1 of Bertsekas,
More information1 Solutions to Tute09
s to Tute0 Questions 4. - 4. are straight forward. Q. 4.4 Show that in a binary tree of N nodes, there are N + NULL pointers. Every node has outgoing pointers. Therefore there are N pointers. Each node,
More informationReinforcement Learning
Reinforcement Learning MDP March May, 2013 MDP MDP: S, A, P, R, γ, µ State can be partially observable: Partially Observable MDPs () Actions can be temporally extended: Semi MDPs (SMDPs) and Hierarchical
More informationLevin Reduction and Parsimonious Reductions
Levin Reduction and Parsimonious Reductions The reduction R in Cook s theorem (p. 266) is such that Each satisfying truth assignment for circuit R(x) corresponds to an accepting computation path for M(x).
More informationEE641 Digital Image Processing II: Purdue University VISE - October 29,
EE64 Digital Image Processing II: Purdue University VISE - October 9, 004 The EM Algorithm. Suffient Statistics and Exponential Distributions Let p(y θ) be a family of density functions parameterized by
More informationNotes on the EM Algorithm Michael Collins, September 24th 2005
Notes on the EM Algorithm Michael Collins, September 24th 2005 1 Hidden Markov Models A hidden Markov model (N, Σ, Θ) consists of the following elements: N is a positive integer specifying the number of
More informationQ1. [?? pts] Search Traces
CS 188 Spring 2010 Introduction to Artificial Intelligence Midterm Exam Solutions Q1. [?? pts] Search Traces Each of the trees (G1 through G5) was generated by searching the graph (below, left) with a
More informationThe Multistep Binomial Model
Lecture 10 The Multistep Binomial Model Reminder: Mid Term Test Friday 9th March - 12pm Examples Sheet 1 4 (not qu 3 or qu 5 on sheet 4) Lectures 1-9 10.1 A Discrete Model for Stock Price Reminder: The
More informationNormal distribution. We say that a random variable X follows the normal distribution if the probability density function of X is given by
Normal distribution The normal distribution is the most important distribution. It describes well the distribution of random variables that arise in practice, such as the heights or weights of people,
More informationComputational Independence
Computational Independence Björn Fay mail@bfay.de December 20, 2014 Abstract We will introduce different notions of independence, especially computational independence (or more precise independence by
More informationInterpolation. 1 What is interpolation? 2 Why are we interested in this?
Interpolation 1 What is interpolation? For a certain function f (x we know only the values y 1 = f (x 1,,y n = f (x n For a point x different from x 1,,x n we would then like to approximate f ( x using
More informationMengdi Wang. July 3rd, Laboratory for Information and Decision Systems, M.I.T.
Practice July 3rd, 2012 Laboratory for Information and Decision Systems, M.I.T. 1 2 Infinite-Horizon DP Minimize over policies the objective cost function J π (x 0 ) = lim N E w k,k=0,1,... DP π = {µ 0,µ
More informationA Theory of Loss-leaders: Making Money by Pricing Below Cost
A Theory of Loss-leaders: Making Money by Pricing Below Cost Maria-Florina Balcan Avrim Blum T-H. Hubert Chan MohammadTaghi Hajiaghayi ABSTRACT We consider the problem of assigning prices to goods of fixed
More informationChapter 7: Estimation Sections
1 / 40 Chapter 7: Estimation Sections 7.1 Statistical Inference Bayesian Methods: Chapter 7 7.2 Prior and Posterior Distributions 7.3 Conjugate Prior Distributions 7.4 Bayes Estimators Frequentist Methods:
More informationCS360 Homework 14 Solution
CS360 Homework 14 Solution Markov Decision Processes 1) Invent a simple Markov decision process (MDP) with the following properties: a) it has a goal state, b) its immediate action costs are all positive,
More informationMonte Carlo and Empirical Methods for Stochastic Inference (MASM11/FMSN50)
Monte Carlo and Empirical Methods for Stochastic Inference (MASM11/FMSN50) Magnus Wiktorsson Centre for Mathematical Sciences Lund University, Sweden Lecture 5 Sequential Monte Carlo methods I January
More information3. Continuous Probability Distributions
3.1 Continuous probability distributions 3. Continuous Probability Distributions K The normal probability distribution A continuous random variable X is said to have a normal distribution if it has a probability
More informationUnobserved Heterogeneity Revisited
Unobserved Heterogeneity Revisited Robert A. Miller Dynamic Discrete Choice March 2018 Miller (Dynamic Discrete Choice) cemmap 7 March 2018 1 / 24 Distributional Assumptions about the Unobserved Variables
More informationExpectations. Definition Let X be a discrete rv with set of possible values D and pmf p(x). The expected value or mean value of X, denoted by E(X ) or
Definition Let X be a discrete rv with set of possible values D and pmf p(x). The expected value or mean value of X, denoted by E(X ) or µ X, is E(X ) = µ X = x D x p(x) Definition Let X be a discrete
More informationMATH 142 Business Mathematics II
MATH 142 Business Mathematics II Summer, 2016, WEEK 2 JoungDong Kim Week 2: 4.1, 4.2, 4.3, 4.4, 4.5 Chapter 4 Rules for the Derivative Section 4.1 Derivatives of Powers, Exponents, and Sums Differentiation
More informationLecture 2. (1) Permanent Income Hypothesis. (2) Precautionary Savings. Erick Sager. September 21, 2015
Lecture 2 (1) Permanent Income Hypothesis (2) Precautionary Savings Erick Sager September 21, 2015 Econ 605: Adv. Topics in Macroeconomics Johns Hopkins University, Fall 2015 Erick Sager Lecture 2 (9/21/15)
More informationMonte Carlo and Empirical Methods for Stochastic Inference (MASM11/FMSN50)
Monte Carlo and Empirical Methods for Stochastic Inference (MASM11/FMSN50) Magnus Wiktorsson Centre for Mathematical Sciences Lund University, Sweden Lecture 6 Sequential Monte Carlo methods II February
More informationBlockchain Economics
Blockchain Economics Joseph Abadi & Markus Brunnermeier (Preliminary and not for distribution) March 9, 2018 Abadi & Brunnermeier Blockchain Economics March 9, 2018 1 / 35 Motivation Ledgers are written
More informationCS188 Spring 2012 Section 4: Games
CS188 Spring 2012 Section 4: Games 1 Minimax Search In this problem, we will explore adversarial search. Consider the zero-sum game tree shown below. Trapezoids that point up, such as at the root, represent
More informationUtility Indifference Pricing and Dynamic Programming Algorithm
Chapter 8 Utility Indifference ricing and Dynamic rogramming Algorithm In the Black-Scholes framework, we can perfectly replicate an option s payoff. However, it may not be true beyond the Black-Scholes
More information6. Genetics examples: Hardy-Weinberg Equilibrium
PBCB 206 (Fall 2006) Instructor: Fei Zou email: fzou@bios.unc.edu office: 3107D McGavran-Greenberg Hall Lecture 4 Topics for Lecture 4 1. Parametric models and estimating parameters from data 2. Method
More informationLecture Note Set 3 3 N-PERSON GAMES. IE675 Game Theory. Wayne F. Bialas 1 Monday, March 10, N-Person Games in Strategic Form
IE675 Game Theory Lecture Note Set 3 Wayne F. Bialas 1 Monday, March 10, 003 3 N-PERSON GAMES 3.1 N-Person Games in Strategic Form 3.1.1 Basic ideas We can extend many of the results of the previous chapter
More informationCh 5. Several Numerical Methods
Ch 5 Several Numerical Methods I Monte Carlo Simulation for Multiple Variables II Confidence Interval and Variance Reduction III Solving Systems of Linear Equations IV Finite Difference Method ( 有限差分法
More informationThe stochastic discount factor and the CAPM
The stochastic discount factor and the CAPM Pierre Chaigneau pierre.chaigneau@hec.ca November 8, 2011 Can we price all assets by appropriately discounting their future cash flows? What determines the risk
More informationDiscrete Random Variables and Probability Distributions. Stat 4570/5570 Based on Devore s book (Ed 8)
3 Discrete Random Variables and Probability Distributions Stat 4570/5570 Based on Devore s book (Ed 8) Random Variables We can associate each single outcome of an experiment with a real number: We refer
More informationAnalysis of Variance and Design of Experiments-II
Analysis of Variance and Design of Experiments-II MODULE I LECTURE - 8 INCOMPLETE BLOCK DESIGNS Dr Shalabh Department of Mathematics & Statistics Indian Institute of Technology Kanpur Generally, we are
More informationSTOCHASTIC INTEGRALS
Stat 391/FinMath 346 Lecture 8 STOCHASTIC INTEGRALS X t = CONTINUOUS PROCESS θ t = PORTFOLIO: #X t HELD AT t { St : STOCK PRICE M t : MG W t : BROWNIAN MOTION DISCRETE TIME: = t < t 1
More informationTop-down particle filtering for Bayesian decision trees
Top-down particle filtering for Bayesian decision trees Balaji Lakshminarayanan 1, Daniel M. Roy 2 and Yee Whye Teh 3 1. Gatsby Unit, UCL, 2. University of Cambridge and 3. University of Oxford Outline
More informationTechniques for Calculating the Efficient Frontier
Techniques for Calculating the Efficient Frontier Weerachart Kilenthong RIPED, UTCC c Kilenthong 2017 Tee (Riped) Introduction 1 / 43 Two Fund Theorem The Two-Fund Theorem states that we can reach any
More informationEcon 6900: Statistical Problems. Instructor: Yogesh Uppal
Econ 6900: Statistical Problems Instructor: Yogesh Uppal Email: yuppal@ysu.edu Lecture Slides 4 Random Variables Probability Distributions Discrete Distributions Discrete Uniform Probability Distribution
More information6.231 DYNAMIC PROGRAMMING LECTURE 3 LECTURE OUTLINE
6.21 DYNAMIC PROGRAMMING LECTURE LECTURE OUTLINE Deterministic finite-state DP problems Backward shortest path algorithm Forward shortest path algorithm Shortest path examples Alternative shortest path
More informationFinancial Economics 4: Portfolio Theory
Financial Economics 4: Portfolio Theory Stefano Lovo HEC, Paris What is a portfolio? Definition A portfolio is an amount of money invested in a number of financial assets. Example Portfolio A is worth
More informationQUANTITATIVE FINANCE RESEARCH CENTRE
QUANTITATIVE FINANCE RESEARCH CENTRE QUANTITATIVE F INANCE RESEARCH CENTRE QUANTITATIVE FINANCE RESEARCH CENTRE Research Paper 288 March 2011 The Evaluation of Multiple Year Gas Sales Agreement with Regime
More informationCOS 513: Gibbs Sampling
COS 513: Gibbs Sampling Matthew Salesi December 6, 2010 1 Overview Concluding the coverage of Markov chain Monte Carlo (MCMC) sampling methods, we look today at Gibbs sampling. Gibbs sampling is a simple
More informationModelling, Estimation and Hedging of Longevity Risk
IA BE Summer School 2016, K. Antonio, UvA 1 / 50 Modelling, Estimation and Hedging of Longevity Risk Katrien Antonio KU Leuven and University of Amsterdam IA BE Summer School 2016, Leuven Module II: Fitting
More informationOptimal Stopping for American Type Options
Optimal Stopping for Department of Mathematics Stockholm University Sweden E-mail: silvestrov@math.su.se ISI 2011, Dublin, 21-26 August 2011 Outline of communication Multivariate Modulated Markov price
More informationA Markovian Futures Market for Computing Power
Fernando Martinez Peter Harrison Uli Harder A distributed economic solution: MaGoG A world peer-to-peer market No central auctioneer Messages are forwarded by neighbours, and a copy remains in their pubs
More information2.1 Mathematical Basis: Risk-Neutral Pricing
Chapter Monte-Carlo Simulation.1 Mathematical Basis: Risk-Neutral Pricing Suppose that F T is the payoff at T for a European-type derivative f. Then the price at times t before T is given by f t = e r(t
More information