Intro Utility theory Utility & Loss Functions Value of Information Summary. Decision Theory. (CIV Machine Learning for Civil Engineers)
|
|
- Julia Owens
- 5 years ago
- Views:
Transcription
1 Decision Theory (CIV654 - Machine Learning for Civil Engineers) Professor: James-A. Goulet Département des génies civil, géologique et des mines Chapter 14 Goulet (219). Probabilistic Machine Learning for Civil Engineers. Chapter 16 Russell, S. and Norvig, P. (1995). Artificial Intelligence, A modern approach. Prentice-Hall. Decision Theory V1.3 Machine Learning for Civil Engineers 1 / 26
2 Intro Utility theory Utility & Loss Functions Value of Information Summary Context Making rational decisions - Fire alarm t 1 min t Pr =.99 t + 1 min Pr =.7.99 Pr =.1.1 Pr =.1.99 Pr = Pr =.9.95 Decision Theory V1.3 Machine Learning for Civil Engineers.95 Polytechnique Montr eal 2 / 26
3 Context Making rational decisions - Soil contamination We have 1 m 3 of soil form an industrial site. What should we do? Pr =.9 $ 1K$ Pr =.1 1$ 1$ E[$ ] = $.9 + 1K$.1 = 1K$ Optimal action E[$ ] = 1$.9 + 1$.1 = 1$ Decision Theory V1.3 Machine Learning for Civil Engineers 3 / 26
4 Nomenclature Nomenclature A = {a 1, a 2,, a A } x Z or R Pr(x) U(a i, x) L(a i, x) U(a i, x) A set of possible actions An outcome in a set of possible states Probability of a state x Utility given a state x and an action a i Loss given a state x and an action a i Decision Theory V1.3 Machine Learning for Civil Engineers 4 / 26
5 Nomenclature Soil contamination example a i {, } {, 1} x {, } {, 1} Pr(x) = [.9,.1] [,, U(a i, x) = U,, [,, L(a i, x) = L,, ] [ ] $ 1K $ 1$ 1$ [ $ ] 1K$ 1$ 1$ ] Decision Theory V1.3 Machine Learning for Civil Engineers 5 / 26
6 Rational decisions Rational decisions Expected utility maximization The perceived benefit of an outcome x i given an action a i is measured by the expected utility or expected loss E[U(a)] = X i=1 U(a, x i) Pr(x i ) E[L(a)] = X i=1 L(a, x i) Pr(x i ) The optimal action a is the one that maximizes the expected utility or minimizes the expected loss a = arg max a E[U(a)] = arg min a E[L(a)] Decision Theory V1.3 Machine Learning for Civil Engineers 6 / 26
7 Intro Utility theory Utility & Loss Functions Value of Information Summary Module #9 Outline Topics organization Supervised learning Decision Making Decision Theory V1.3 Machine Learning for Civil Engineers 1 Revision probability & linear algebra µ σ ( Bayesian Estimation 4 FX(x) 2 x 2 µ σ µ µ+σ x 2 4 p(b A)p(A) p(b) p(a B) = s = MCMC sampling & Newton Regression θ µ µ+σ f X(x).4 2 Probability distributions Introduction θ1 5 θ θ1 Patogens [ppm] Machine Learning Basics ( [ ] cl+ d= fx D(x d) Probability theory Classification d=1 d=2 d=3 d=4 d=5 1 P(D S) Intro Utility theory Utility & Loss Functions Value of Information.2.4 S={4,5} 1 x.6 S={2,3}.8 1 S={,1} State-space model for time-series E[U] = 3.7 E[U] = 7. 8 Decision Theory 9 AI & Sequential decision problems Polytechnique Montr eal 7 / 26
8 Section Outline Utility theory 2.1 Lotteries 2.2 Axioms of utility theory Decision Theory V1.3 Machine Learning for Civil Engineers 8 / 26
9 Lotteries Nomenclature for ordering preferences A lottery: L i = [{p 1, x 1 }, {p 2, x 2 },, {p X, x X }] L = [{1.,, }{.,, }] L = [{.9,, }{.1,, }] A decision maker L i L j L i L j L i L j prefers L i over L j is indifferent between L i and L j prefers L i over L j or is indifferent Decision Theory V1.3 Machine Learning for Civil Engineers 9 / 26
10 Axioms of utility theory Axioms of utility theory What is defining a rational behaviour? Orderability: Exactly one of (L i L j ), (L j L i ), (L i L j ) holds Transitivity: if (L i L j ) (L j L k ), then (L i L k ) Continuity: if (L i L j L k ), then p : [{p, L i }; {1 p, L k }] L j Substitutability: if (L i L j ), then [{p, L i }; {1 p, L k }] [{p, L j }; {1 p, L k }] Monotonicity: if L i L j, then (p q [{p, L i }; {1 p, L j }] [{q, L i }; {1 q, L j }]) Decomposability:...no fun in gambling Decision Theory V1.3 Machine Learning for Civil Engineers 1 / 26
11 Section Outline Utility & Loss Functions 3.1 Utility 3.2 Non-linear utility functions 3.3 Utility and Loss functions U(v) & L(v) 3.4 E[L(v(a i, X ))] and risk aversion Decision Theory V1.3 Machine Learning for Civil Engineers 11 / 26
12 Utility Axioms utility Existence of a utility function: U(L i ) U(L j ) L i L j U(L i ) = U(L j ) L i L j Expected utility of a lottery: E[U([{p 1, x 1 }, {p 2, x 2 },, {p X, x X }])] = X p i U(x i ) i=1 Decision Theory V1.3 Machine Learning for Civil Engineers 12 / 26
13 Non-linear utility functions Do you want to take the lottery? L = [{ 1 2, + }, { 1 2, - }] L = [{1, +$}] Which lottery do you choose? Why? E[$(L )] = $ $ = +5$ E[$(L )] = $ Are you being irrational? For individuals U($) and L($) are non-linear... Decision Theory V1.3 Machine Learning for Civil Engineers 13 / 26
14 1Utility, (v) Utility and Loss functions U(v) & L(v) Risk aversion and utility functions U(v) U(v): An utility function weight monetary value (v) as a function of risk aversion/propension (± 1$ not the same effet if you have 1$ or 1M$) Risk seeking Risk neutral Risk averse v U(v) = v k People/organizations are risk averse k > 1 Risk seeking k = 1 Neutral < k < 1 Risk averse Decision Theory V1.3 Machine Learning for Civil Engineers 14 / 26
15 1Loss, (v) Utility and Loss functions U(v) & L(v) Risk aversion and loss functions L(v) L(v): A loss function weight monetary value (v) as a function of risk aversion/propension Risk seeking Risk neutral Risk averse v L(v) = v k People/organizations are risk averse k > 1 Risk averse k = 1 Neutral < k < 1 Risk seeking Decision Theory V1.3 Machine Learning for Civil Engineers 15 / 26
16 1Loss, (v) Utility and Loss functions U(v) & L(v) Attitude toward risks Risk seeking Risk neutral Risk averse v k > 1 Risk averse L(v) = v k k = 1 Neutral < k < 1 Risk seeking A neutral attitude toward risks minimizes the expected costs over a multiple decisions Insurance compagnies: neutral attitude toward risks Insured people: risk averse; they pay a premium not to be in a risk neutral position (i.e. expected costs are higher over multiple decisions) Decision Theory V1.3 Machine Learning for Civil Engineers 16 / 26
17 1Loss, (v) Utility and Loss functions U(v) & L(v) Expected Loss Risk seeking Risk neutral Risk averse v Value Loss: Value x = x = a = v(, ) v(, ) a = v(, ) v(, ) Loss x = x = E[L(v(a, X ))] a = L(v(, )) L(v(, )) E[L(v(, X ))] a = L(v(, )) L(v(, )) E[L(v(, X ))] Decision Theory V1.3 Machine Learning for Civil Engineers 17 / 26
18 E[L(v(a i, X ))] and risk aversion E[L(v(a i, X ))] and risk aversion (ex. discrete) [ ] 1 E[L(v(a1, X)] = E[L(v(a2, X)] Loss, L(v) 1 p(l(v) a i, x) Action a 1 Action a 2 p(v ai, x) 1 v(a i, x) 1.5 E[v(a1, X)] = E[v(a2, X)].5 1 v(a i, x) Risk perception neutral: E[L(v(a 1, X ))] E[L(v(a 2, X ))] [CIV ML/Decision/PCgA discrete.m] Decision Theory V1.3 Machine Learning for Civil Engineers 18 / 26
19 E[L(v(a i, X ))] and risk aversion E[L(v(a i, X ))] and risk aversion (ex. discrete) [ ] 1 E[L(v(a1, X)] E[L(v(a2, X)] p(l(v) a i, x) Action a 1 Action a 2 p(v ai, x) Loss, L(v) 1 1 v(a i, x) 1.5 E[v(a1, X)] = E[v(a2, X)].5 1 v(a i, x) Risk perception neutral: E[L(v(a 1, X ))] E[L(v(a 2, X ))] [CIV ML/Decision/PCgA discrete.m] Decision Theory V1.3 Machine Learning for Civil Engineers 18 / 26
20 E[L(v(a i, X ))] and risk aversion E[L(v(a i, X ))] and risk aversion (ex. continuous) [ ] E[L(v(a1, X))] = E[L(v(a2, X))] 1 Loss, L(v) f(l(v) a i, x) 1 v(a i, x) Action a 1 Action a 2 f(v ai, x) E[v(a1, X)] = E[v(a2, X)].5 1 v(a i, x) Risk perception neutral: E[L(v(a 1, X ))] E[L(v(a 2, X ))] [CIV ML/Decision/PCgA continuous.m] Decision Theory V1.3 Machine Learning for Civil Engineers 19 / 26
21 E[L(v(a i, X ))] and risk aversion E[L(v(a i, X ))] and risk aversion (ex. continuous) [ ] 1 E[L(v(a 2, X))] E[L(v(a 1, X))] f(l(v) a i, x) Loss, L(v) 1 v(a i, x) Action a 1 Action a 2 f(v ai, x) E[v(a1, X)] = E[v(a2, X)].5 1 v(a i, x) Risk perception neutral: E[L(v(a 1, X ))] E[L(v(a 2, X ))] [CIV ML/Decision/PCgA continuous.m] Decision Theory V1.3 Machine Learning for Civil Engineers 19 / 26
22 Section Outline Value of Information 4.1 Value of perfect information 4.2 Value of imperfect information 4.3 Exemples Decision Theory V1.3 Machine Learning for Civil Engineers 2 / 26
23 Value of perfect information Expected utility of collecting information In cases where the value of a state x is imperfectly known,one possible action is to collect information about X. X E[U(a )] = max U(a, x i ) Pr(x i ) a U(a, x = y) = max U(a, x = y) a Because y has not been observed yet, we must consider all possibilities Y = X i according to their probability E[U(ã )] = X i=1 Value of perfect information i=1 max [U(a, x i )] Pr(x i ) a VPI (y) = E[U(ã )] E[U(a )] Decision Theory V1.3 Machine Learning for Civil Engineers 21 / 26
24 Value of perfect information Soil contamination example L(a, x) x = x = a = 1$ 1$ a = $ 1K$ Current expected costs conditional on actions E[L(, X )] = $.9 + 1K$.1 = 1K$ E[L(, X )] = 1$.9 + 1$.1 = 1$ = E[L(a, X )] Expected costs conditional on perfect information E[L(ã )] = X i=1 min a (L(a, x i)) Pr(x i ) = $.9 }{{} y=x= + 1$.1 }{{} y=x= = 1$ Value of perfect information VPI (y) = E[L(a )] E[L(ã )] = 9$ Decision Theory V1.3 Machine Learning for Civil Engineers 22 / 26
25 Value of perfect information Value of information The value of information represents how much you are willing to pay for an information. What if the information is not perfect? Decision Theory V1.3 Machine Learning for Civil Engineers 23 / 26
26 Value of imperfect information Value of imperfect information Discrete state case E[U(ã )] = X X i=1 j=1 max a (U(a, y i)) Pr(y i x j ) Pr(x j ) Continuous state case E[U(ã )] = max(u(a, y)) f (y x) f (x)dydx a Decision Theory V1.3 Machine Learning for Civil Engineers 24 / 26
27 Exemples Soil contamination example - Discrete case L(a, x) x = x = a = 1$ 1$ a = $ 1K$ Pr(y = ) = 1 Pr(y = ) =.95 Current expected costs E[L(a, X )] = 1$ Expected costs conditional on imperfect information E[L(ã )] = X i=1 X j=1 min a (L(a, y i)) Pr(y i x j ) Pr(x j ) = $.9 }{{} x=y= = 59.5$ Value of information + ( 1$.95 }{{} y= + 1K$.5 ).1 }{{} y= } {{ } x= VOI (y) = E[L(a )] E[L(ã )] = 4.5$ Decision Theory V1.3 Machine Learning for Civil Engineers 25 / 26
28 Summary Rational Decision: Choose the action ai which minimize the expected loss L(a, x) or maximizes the expected cost U(a, x) a i = arg min a i E[L(a i, x)] = arg max E[U(a i, x)] a i L(v(a, x)) & U(v(a, x)): Subjective weight on value as a function of the attitude toward risks (± 1$ not the same effect if you have 1$ or 1M$) 1Loss, (v) Risk seeking Risk neutral Risk averse Value of information: Value you should be willing to pay for information VOI (y) = E[U(a )] E[U(ã )] Value of perfect information: E[U(ã )] = X i=1 max [U(a, x a i )] Pr(x i ) Value of imperfect information: v k > 1 Risk averse L(v) = v k k = 1 Neutral < k < 1 Risk seeking E[U(ã )] = X X max a (U(a, y)) Pr(y x j ) Pr(x j ) i=1 j=1 Decision Theory V1.3 Machine Learning for Civil Engineers 26 / 26
Artificial Intelligence & Sequential Decision Problems
Artificial Intelligence & Sequential Decision Problems (CIV6540 - Machine Learning for Civil Engineers) Professor: James-A. Goulet Département des génies civil, géologique et des mines Chapter 15 Goulet
More information12735: Urban Systems Modeling. Loss and decisions. instructor: Matteo Pozzi. Lec : Urban Systems Modeling Lec. 11 Loss and decisions
1735: Urban Systems Modeling Lec. 11 Loss and decisions instructor: Matteo Pozzi 1 outline introduction example of decision under uncertainty attitude toward risk principle of minimum expected loss Bayesian
More informationChoice under Uncertainty
In the Name of God Sharif University of Technology Graduate School of Management and Economics Microeconomics 2 44706 (1394-95 2 nd term) Group 2 Dr. S. Farshad Fatemi Chapter 6: Choice under Uncertainty
More informationDecision Theory Intro: Preferences and Utility
Decision Theory Intro: Preferences and Utility CPSC 322 Lecture 29 March 22, 2006 Textbook 9.5 Decision Theory Intro: Preferences and Utility CPSC 322 Lecture 29, Slide 1 Lecture Overview Recap Decision
More informationLecture Note Set 4 4 UTILITY THEORY. IE675 Game Theory. Wayne F. Bialas 1 Thursday, March 6, Introduction
IE675 Game Theory Lecture Note Set 4 Wayne F. Bialas 1 Thursday, March 6, 2003 4 UTILITY THEORY 4.1 Introduction 4.2 The basic theory This section is really independent of the field of game theory, and
More informationLast update: April 15, Rational decisions. CMSC 421: Chapter 16. CMSC 421: Chapter 16 1
Last update: April 15, 2010 Rational decisions CMSC 421: Chapter 16 CMSC 421: Chapter 16 1 Outline Rational preferences Utilities Money Multiattribute utilities Decision networks Value of information CMSC
More informationIntroduction Linear Regression Gaussian Process Regression (GPR) Examples Adv. topics Summary. Regression methods
Regression methods (CIV654 - Machine Learning for Civil Engineers) Professor: James-A. Goulet Département des génies civil, géologique et des mines Chapters 8 Goulet (218). Probabilistic Machine Learning
More informationHomework #6 (10/18/2017)
Homework #6 (0/8/207). Let G be the set of compound gambles over a finite set of deterministic payoffs {a, a 2,...a n } R +. A decision maker s preference relation over compound gambles can be represented
More informationCS 798: Multiagent Systems
CS 798: Multiagent Systems and Utility Kate Larson Cheriton School of Computer Science University of Waterloo January 6, 2010 Outline 1 Self-Interested Agents 2 3 4 5 Self-Interested Agents We are interested
More informationRational decisions From predicting the effect to optimal intervention
Rational decisions From predicting the effect to optimal intervention ADAPTED FROM AIMA SLIDES Russel&Norvig:Artificial Intelligence: a modern approach AIMA: Rational decisions AIMA: Rational decisions
More informationAlmost essential: Consumption and Uncertainty Probability Distributions MICROECONOMICS
Prerequisites Almost essential: Consumption and Uncertainty Probability Distributions RISK MICROECONOMICS Principles and Analysis Frank Cowell July 2017 1 Risk and uncertainty In dealing with uncertainty
More informationEvaluation for Pacman. CS 188: Artificial Intelligence Fall Iterative Deepening. α-β Pruning Example. α-β Pruning Pseudocode.
CS 188: Artificial Intelligence Fall 2008 Evaluation for Pacman Lecture 7: Expectimax Search 9/18/2008 [DEMO: thrashing, smart ghosts] Dan Klein UC Berkeley Many slides over the course adapted from either
More informationCS 188: Artificial Intelligence Fall 2008
CS 188: Artificial Intelligence Fall 2008 Lecture 7: Expectimax Search 9/18/2008 Dan Klein UC Berkeley Many slides over the course adapted from either Stuart Russell or Andrew Moore 1 1 Evaluation for
More informationBasic Probability and Decisions
Basic Probability and Decisions Chris Amato Northeastern University Some images and slides are used from: Rob Platt, CS188 UC Berkeley, AIMA Uncertainty Let action A t = leave for airport t minutes before
More information1 Uncertainty. These notes correspond to chapter 2 of Jehle and Reny.
These notes correspond to chapter of Jehle and Reny. Uncertainty Until now we have considered our consumer s making decisions in a world with perfect certainty. However, we can extend the consumer theory
More informationRational preferences. Rational decisions. Outline. Rational preferences contd. Maximizing expected utility. Preferences
Rational preferences Rational decisions hapter 16 Idea: preferences of a rational agent must obey constraints. Rational preferences behavior describable as maximization of expected utility onstraints:
More informationRecitation 7: Uncertainty. Xincheng Qiu
Econ 701A Fall 2018 University of Pennsylvania Recitation 7: Uncertainty Xincheng Qiu (qiux@sas.upenn.edu 1 Expected Utility Remark 1. Primitives: in the basic consumer theory, a preference relation is
More informationTowards a Theory of Decision Making without Paradoxes
Towards a Theory of Decision Making without Paradoxes Roman V. Belavkin (R.Belavkin@mdx.ac.uk) School of Computing Science, Middlesex University London NW4 4BT, United Kingdom Abstract Human subjects often
More informationLecture 14: Introduction to Decision Making
Lecture 14: Introduction to Decision Making Preferences Utility functions Maximizing exected utility Value of information Actions and consequences So far, we have focused on ways of modeling a stochastic,
More informationQUANTIFYING THE ECONOMIC VALUE OF WEATHER FORECASTS: REVIEW OF METHODS AND RESULTS
QUANTIFYING THE ECONOMIC VALUE OF WEATHER FORECASTS: REVIEW OF METHODS AND RESULTS Rick Katz Institute for Study of Society and Environment National Center for Atmospheric Research Boulder, CO USA Email:
More informationBayesian Networks BY: MOHAMAD ALSABBAGH
Bayesian Networks BY: MOHAMAD ALSABBAGH Outlines Introduction Bayes Rule Bayesian Networks (BN) Representation Size of a Bayesian Network Inference via BN BN Learning Dynamic BN Introduction Conditional
More informationIntroduction to Systems Analysis and Decision Making Prepared by: Jakub Tomczak
Introduction to Systems Analysis and Decision Making Prepared by: Jakub Tomczak 1 Introduction. Random variables During the course we are interested in reasoning about considered phenomenon. In other words,
More informationUtility and Decision Making 1
Utility and Decision Making 1 Suppose that a casino offers to keep flipping a coin until it comes up heads for the first time. If it comes up heads on the 1st flip, you win $1. If it comes up heads for
More informationReview of topics since what was covered in the midterm: Topics that we covered before the midterm (also may be included in final):
Review of topics since what was covered in the midterm: Subgame-perfect eqms in extensive games with perfect information where players choose a number (first-order conditions, boundary conditions, favoring
More informationDerivatives in 2D. Outline. James K. Peterson. November 9, Derivatives in 2D! Chain Rule
Derivatives in 2D James K. Peterson Department of Biological Sciences and Department of Mathematical Sciences Clemson University November 9, 2016 Outline Derivatives in 2D! Chain Rule Let s go back to
More informationInformation, Utility & Bounded Rationality
Information, Utility & Bounded Rationality Pedro A. Ortega and Daniel A. Braun Department of Engineering, University of Cambridge Trumpington Street, Cambridge, CB2 PZ, UK {dab54,pao32}@cam.ac.uk Abstract.
More informationMATH 446/546 Homework 2: Due October 8th, 2014
MATH 446/546 Homework 2: Due October 8th, 2014 Answer the following questions. Some of which come from Winston s text book. 1. We are going to invest $1,000 for a period of 6 months. Two potential investments
More informationProbability and Decision Theory
Probability and Decision Theory Robert Platt Northeastern University Some images and slides are used from: 1. AIMA 2. Chris Amato 3. Stacy Marsella QUANTIFYING UNCERTAINTY WITH PROBABILITIES Generally
More informationNotes on Decision Theory and Prediction
Notes on Decision Theory and Prediction Ronald Christensen Professor of Statistics Department of Mathematics and Statistics University of New Mexico October 7, 2014 1. Decision Theory Decision theory is
More informationFun with Utilities. 1 Introduction. 2 Preference Patterns. Michael A. Goodrich. September 1, 2005
Fun with Utilities Michael A. Goodrich September 1, 2005 1 Introduction In these notes, we will introduce the axioms underlying utility theory, and then prove one half of what I call the Utility Theorem.
More informationRational decisions. Chapter 16. Chapter 16 1
Rational decisions Chapter 16 Chapter 16 1 Outline Rational preferences Utilities Money Multiattribute utilities Decision networks Value of information Chapter 16 2 Preferences An agent chooses among prizes
More informationBayesian Networks. instructor: Matteo Pozzi. x 1. x 2. x 3 x 4. x 5. x 6. x 7. x 8. x 9. Lec : Urban Systems Modeling
12735: Urban Systems Modeling Lec. 09 Bayesian Networks instructor: Matteo Pozzi x 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8 x 9 1 outline example of applications how to shape a problem as a BN complexity of the inference
More informationProbabilistic Representation and Reasoning
Probabilistic Representation and Reasoning Alessandro Panella Department of Computer Science University of Illinois at Chicago May 4, 2010 Alessandro Panella (CS Dept. - UIC) Probabilistic Representation
More informationBayesian Networks Inference with Probabilistic Graphical Models
4190.408 2016-Spring Bayesian Networks Inference with Probabilistic Graphical Models Byoung-Tak Zhang intelligence Lab Seoul National University 4190.408 Artificial (2016-Spring) 1 Machine Learning? Learning
More informationTwo-Stage-Partitional Representation and Dynamic Consistency 1
Two-Stage-Partitional Representation and Dynamic Consistency 1 Suguru Ito June 5 th, 2015 Abstract In this paper, we study an individual who faces a three-period decision problem when she accepts the partitional
More informationThis corresponds to a within-subject experiment: see same subject make choices from different menus.
Testing Revealed Preference Theory, I: Methodology The revealed preference theory developed last time applied to a single agent. This corresponds to a within-subject experiment: see same subject make choices
More informationY. Xiang, Inference with Uncertain Knowledge 1
Inference with Uncertain Knowledge Objectives Why must agent use uncertain knowledge? Fundamentals of Bayesian probability Inference with full joint distributions Inference with Bayes rule Bayesian networks
More informationA Study in Preference Elicitation under Uncertainty
A Study in Preference Elicitation under Uncertainty by Greg Hines A thesis presented to the University of Waterloo in fulfillment of the thesis requirement for the degree of Doctor of Philosophy in Computer
More informationDepartment of Economics, University of California, Davis Ecn 200C Micro Theory Professor Giacomo Bonanno ANSWERS TO PRACTICE PROBLEMS 18
Department of Economics, University of California, Davis Ecn 00C Micro Theory Professor Giacomo Bonanno ANSWERS TO PRACTICE PROBEMS 8. If price is Number of cars offered for sale Average quality of cars
More informationECO 317 Economics of Uncertainty Fall Term 2009 Notes for lectures 3. Risk Aversion
Reminders ECO 317 Economics of Uncertainty Fall Term 009 Notes for lectures 3. Risk Aversion On the space of lotteries L that offer a finite number of consequences (C 1, C,... C n ) with probabilities
More information14.12 Game Theory Lecture Notes Theory of Choice
14.12 Game Theory Lecture Notes Theory of Choice Muhamet Yildiz (Lecture 2) 1 The basic theory of choice We consider a set X of alternatives. Alternatives are mutually exclusive in the sense that one cannot
More information1 Uncertainty and Insurance
Uncertainty and Insurance Reading: Some fundamental basics are in Varians intermediate micro textbook (Chapter 2). A good (advanced, but still rather accessible) treatment is in Kreps A Course in Microeconomic
More informationUncertainty & Decision
Uncertainty & Decision von Neumann Morgenstern s Theorem Stéphane Airiau & Umberto Grandi ILLC - University of Amsterdam Stéphane Airiau & Umberto Grandi (ILLC) - Uncertainty & Decision von Neumann Morgenstern
More informationApril 9, Depto. de Ing. de Sistemas e Industrial Universidad Nacional de Colombia, Bogotá. Linear Classification Models. Fabio A. González Ph.D.
Depto. de Ing. de Sistemas e Industrial Universidad Nacional de Colombia, Bogotá April 9, 2018 Content 1 2 3 4 Outline 1 2 3 4 problems { C 1, y(x) threshold predict(x) = C 2, y(x) < threshold, with threshold
More informationAnscombe & Aumann Expected Utility Betting and Insurance
Anscombe & Aumann Expected Utility Betting and Insurance Econ 2100 Fall 2017 Lecture 11, October 3 Outline 1 Subjective Expected Utility 2 Qualitative Probabilities 3 Allais and Ellsebrg Paradoxes 4 Utility
More informationProbabilistic Graphical Models for Image Analysis - Lecture 1
Probabilistic Graphical Models for Image Analysis - Lecture 1 Alexey Gronskiy, Stefan Bauer 21 September 2018 Max Planck ETH Center for Learning Systems Overview 1. Motivation - Why Graphical Models 2.
More informationComputational Genomics
Computational Genomics http://www.cs.cmu.edu/~02710 Introduction to probability, statistics and algorithms (brief) intro to probability Basic notations Random variable - referring to an element / event
More information16.4 Multiattribute Utility Functions
285 Normalized utilities The scale of utilities reaches from the best possible prize u to the worst possible catastrophe u Normalized utilities use a scale with u = 0 and u = 1 Utilities of intermediate
More informationMachine Learning and Data Mining. Bayes Classifiers. Prof. Alexander Ihler
+ Machine Learning and Data Mining Bayes Classifiers Prof. Alexander Ihler A basic classifier Training data D={x (i),y (i) }, Classifier f(x ; D) Discrete feature vector x f(x ; D) is a con@ngency table
More informationIntroduction to Machine Learning. Maximum Likelihood and Bayesian Inference. Lecturers: Eran Halperin, Yishay Mansour, Lior Wolf
1 Introduction to Machine Learning Maximum Likelihood and Bayesian Inference Lecturers: Eran Halperin, Yishay Mansour, Lior Wolf 2013-14 We know that X ~ B(n,p), but we do not know p. We get a random sample
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 informationIntroduction to Bayesian Statistics
School of Computing & Communication, UTS January, 207 Random variables Pre-university: A number is just a fixed value. When we talk about probabilities: When X is a continuous random variable, it has a
More informationRational decisions. Chapter 16 1
Rational decisions Chapter 16 Chapter 16 1 Rational preferences Utilities Money Multiattribute utilities Decision networks Value of information Outline Chapter 16 2 Preferences An agent chooses among prizes
More informationIntro to Probability. Andrei Barbu
Intro to Probability Andrei Barbu Some problems Some problems A means to capture uncertainty Some problems A means to capture uncertainty You have data from two sources, are they different? Some problems
More informationChapter 6: Rational Expr., Eq., and Functions Lecture notes Math 1010
Section 6.1: Rational Expressions and Functions Definition of a rational expression Let u and v be polynomials. The algebraic expression u v is a rational expression. The domain of this rational expression
More informationMAS368: Decision making and Bayes linear methods
MAS368: Decision making and Bayes linear methods Module leader: Dr Darren J Wilkinson email: d.j.wilkinson@ncl.ac.uk Office: M515, Phone: 7320 Every day, we are all faced with the problem of having to
More informationSYDE 372 Introduction to Pattern Recognition. Probability Measures for Classification: Part I
SYDE 372 Introduction to Pattern Recognition Probability Measures for Classification: Part I Alexander Wong Department of Systems Design Engineering University of Waterloo Outline 1 2 3 4 Why use probability
More informationMachine Learning for Large-Scale Data Analysis and Decision Making A. Week #1
Machine Learning for Large-Scale Data Analysis and Decision Making 80-629-17A Week #1 Today Introduction to machine learning The course (syllabus) Math review (probability + linear algebra) The future
More informationIntroduction to Machine Learning. Lecture 2
Introduction to Machine Learning Lecturer: Eran Halperin Lecture 2 Fall Semester Scribe: Yishay Mansour Some of the material was not presented in class (and is marked with a side line) and is given for
More informationIntroduction to Game Theory: Simple Decisions Models
Introduction to Game Theory: Simple Decisions Models John C.S. Lui Department of Computer Science & Engineering The Chinese University of Hong Kong John C.S. Lui (CUHK) Advanced Topics in Network Analysis
More information3 Intertemporal Risk Aversion
3 Intertemporal Risk Aversion 3. Axiomatic Characterization This section characterizes the invariant quantity found in proposition 2 axiomatically. The axiomatic characterization below is for a decision
More informationLecture 9: Naive Bayes, SVM, Kernels. Saravanan Thirumuruganathan
Lecture 9: Naive Bayes, SVM, Kernels Instructor: Outline 1 Probability basics 2 Probabilistic Interpretation of Classification 3 Bayesian Classifiers, Naive Bayes 4 Support Vector Machines Probability
More informationPreference, Choice and Utility
Preference, Choice and Utility Eric Pacuit January 2, 205 Relations Suppose that X is a non-empty set. The set X X is the cross-product of X with itself. That is, it is the set of all pairs of elements
More informationSkewed Noise. David Dillenberger 1 Uzi Segal 2. 1 University of Pennsylvania 2 Boston College and WBS
Skewed Noise David Dillenberger 1 Uzi Segal 2 1 University of Pennsylvania 2 Boston College and WBS Introduction Compound lotteries (lotteries over lotteries over outcomes): 1 1 4 3 4 1 2 1 2 1 2 1 2 1
More informationMODULE 10 Bayes Classifier LESSON 20
MODULE 10 Bayes Classifier LESSON 20 Bayesian Belief Networks Keywords: Belief, Directed Acyclic Graph, Graphical Model 1 Bayesian Belief Network A Bayesian network is a graphical model of a situation
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 informationOptimal Illusion of Control
Optimal Illusion of Control and related perception biases Olivier Gossner 1 Jakub Steiner 2 1 École Polytechnique 2 Cerge-ei and Edinburgh Evanston June 2016 1/36 predictable errors in perception of payoff-relevant
More informationMachine Learning and Data Mining. Linear classification. Kalev Kask
Machine Learning and Data Mining Linear classification Kalev Kask Supervised learning Notation Features x Targets y Predictions ŷ = f(x ; q) Parameters q Program ( Learner ) Learning algorithm Change q
More informationAn Introduction to Statistical and Probabilistic Linear Models
An Introduction to Statistical and Probabilistic Linear Models Maximilian Mozes Proseminar Data Mining Fakultät für Informatik Technische Universität München June 07, 2017 Introduction In statistical learning
More informationChoice Theory. Matthieu de Lapparent
Choice Theory Matthieu de Lapparent matthieu.delapparent@epfl.ch Transport and Mobility Laboratory, School of Architecture, Civil and Environmental Engineering, Ecole Polytechnique Fédérale de Lausanne
More informationMath for Machine Learning Open Doors to Data Science and Artificial Intelligence. Richard Han
Math for Machine Learning Open Doors to Data Science and Artificial Intelligence Richard Han Copyright 05 Richard Han All rights reserved. CONTENTS PREFACE... - INTRODUCTION... LINEAR REGRESSION... 4 LINEAR
More informationEE290O / IEOR 290 Lecture 05
EE290O / IEOR 290 Lecture 05 Roy Dong September 7, 2017 In this section, we ll cover one approach to modeling human behavior. In this approach, we assume that users pick actions that maximize some function,
More informationMath 10C - Fall Final Exam
Math 1C - Fall 217 - Final Exam Problem 1. Consider the function f(x, y) = 1 x 2 (y 1) 2. (i) Draw the level curve through the point P (1, 2). Find the gradient of f at the point P and draw the gradient
More informationChapter 03: Bayesian Networks
LEARNING AND INFERENCE IN GRAPHICAL MODELS Chapter 03: Bayesian Networks Dr. Martin Lauer University of Freiburg Machine Learning Lab Karlsruhe Institute of Technology Institute of Measurement and Control
More informationMODULE -4 BAYEIAN LEARNING
MODULE -4 BAYEIAN LEARNING CONTENT Introduction Bayes theorem Bayes theorem and concept learning Maximum likelihood and Least Squared Error Hypothesis Maximum likelihood Hypotheses for predicting probabilities
More informationBayesian RL Seminar. Chris Mansley September 9, 2008
Bayesian RL Seminar Chris Mansley September 9, 2008 Bayes Basic Probability One of the basic principles of probability theory, the chain rule, will allow us to derive most of the background material in
More information4.1. Chapter 4. timing risk information utility
4. Chapter 4 timing risk information utility 4.2. Money pumps Rationality in economics is identified with consistency. In particular, a preference relation must be total and transitive Transitivity implies
More informationMathematics for Economics ECON MA/MSSc in Economics-2017/2018. Dr. W. M. Semasinghe Senior Lecturer Department of Economics
Mathematics for Economics ECON 53035 MA/MSSc in Economics-2017/2018 Dr. W. M. Semasinghe Senior Lecturer Department of Economics MATHEMATICS AND STATISTICS LERNING OUTCOMES: By the end of this course unit
More information6.262: Discrete Stochastic Processes 2/2/11. Lecture 1: Introduction and Probability review
6.262: Discrete Stochastic Processes 2/2/11 Lecture 1: Introduction and Probability review Outline: Probability in the real world Probability as a branch of mathematics Discrete stochastic processes Processes
More informationWeak* Axiom of Independence and the Non-Expected Utility Theory
Review of Economic Analysis 4 (01) 67-75 1973-3909/01067 Weak* Axiom of Independence and the Non-Expected Utility Theory TAPAN BISWAS University of Hull 1 Introduction The axiomatic foundation of the expected
More informationProbabilistic Graphical Models
Probabilistic Graphical Models Introduction. Basic Probability and Bayes Volkan Cevher, Matthias Seeger Ecole Polytechnique Fédérale de Lausanne 26/9/2011 (EPFL) Graphical Models 26/9/2011 1 / 28 Outline
More informationCertainty Equivalent Representation of Binary Gambles That Are Decomposed into Risky and Sure Parts
Review of Economics & Finance Submitted on 28/Nov./2011 Article ID: 1923-7529-2012-02-65-11 Yutaka Matsushita Certainty Equivalent Representation of Binary Gambles That Are Decomposed into Risky and Sure
More informationBasic Game Theory. Kate Larson. January 7, University of Waterloo. Kate Larson. What is Game Theory? Normal Form Games. Computing Equilibria
Basic Game Theory University of Waterloo January 7, 2013 Outline 1 2 3 What is game theory? The study of games! Bluffing in poker What move to make in chess How to play Rock-Scissors-Paper Also study of
More informationGame Theory. Georg Moser. Institute of Computer UIBK. Winter GM (Institute of Computer UIBK) Game Theory 1/19 Organisation
OLC mputational gic Game Theory Georg Moser Institute of Computer Science @ UIBK Winter 2008 GM (Institute of Computer Science @ UIBK) Game Theory 1/19 Organisation Organisation Time and Place Thursday,
More informationComments on prospect theory
Comments on prospect theory Abstract Ioanid Roşu This note presents a critique of prospect theory, and develops a model for comparison of two simple lotteries, i.e. of the form ( x 1, p1; x 2, p 2 ;...;
More information2534 Lecture 2: Utility Theory
2534 Lecture 2: Utility Theory Tutorial on Bayesian Networks: Weds, Sept.17, 5-6PM, PT266 LECTURE ORDERING: Game Theory before MDPs? Or vice versa? Preference orderings Decision making under strict uncertainty
More informationNaïve Bayes classification
Naïve Bayes classification 1 Probability theory Random variable: a variable whose possible values are numerical outcomes of a random phenomenon. Examples: A person s height, the outcome of a coin toss
More informationValue of Information Analysis with Structural Reliability Methods
Accepted for publication in Structural Safety, special issue in the honor of Prof. Wilson Tang August 2013 Value of Information Analysis with Structural Reliability Methods Daniel Straub Engineering Risk
More informationRationality and Uncertainty
Rationality and Uncertainty Based on papers by Itzhak Gilboa, Massimo Marinacci, Andy Postlewaite, and David Schmeidler Warwick Aug 23, 2013 Risk and Uncertainty Dual use of probability: empirical frequencies
More informationQuantum Probability in Cognition. Ryan Weiss 11/28/2018
Quantum Probability in Cognition Ryan Weiss 11/28/2018 Overview Introduction Classical vs Quantum Probability Brain Information Processing Decision Making Conclusion Introduction Quantum probability in
More informationRecursive Ambiguity and Machina s Examples
Recursive Ambiguity and Machina s Examples David Dillenberger Uzi Segal May 0, 0 Abstract Machina (009, 0) lists a number of situations where standard models of ambiguity aversion are unable to capture
More informationWho Learns Better Bayesian Network Structures
Who Learns Better Bayesian Network Structures Constraint-Based, Score-based or Hybrid Algorithms? Marco Scutari 1 Catharina Elisabeth Graafland 2 José Manuel Gutiérrez 2 1 Department of Statistics, UK
More informationTransitive Regret over Statistically Independent Lotteries
Transitive Regret over Statistically Independent Lotteries April 2, 2012 Abstract Preferences may arise from regret, i.e., from comparisons with alternatives forgone by the decision maker. We show that
More information* Professor of Finance at INSEAD, Boulevard de Constance, Fontainebleau Cedex, France.
DIFFERENTIABLE UTILITY by L. T. NIELSEN* 97/28/FIN * Professor of Finance at INSEAD, Boulevard de Constance, Fontainebleau 77305 Cedex, France. A working paper in the INSEAD Working Paper Series is intended
More informationAarti Singh. Lecture 2, January 13, Reading: Bishop: Chap 1,2. Slides courtesy: Eric Xing, Andrew Moore, Tom Mitchell
Machine Learning 0-70/5 70/5-78, 78, Spring 00 Probability 0 Aarti Singh Lecture, January 3, 00 f(x) µ x Reading: Bishop: Chap, Slides courtesy: Eric Xing, Andrew Moore, Tom Mitchell Announcements Homework
More informationSystems Engineering - Decision Analysis in Engineering Michael Havbro Faber Aalborg University, Denmark
Risk and Safety Management Aalborg University, Esbjerg, 2017 Systems Engineering - Decision Analysis in Engineering Michael Havbro Faber Aalborg University, Denmark 1/50 M. H. Faber Systems Engineering,
More informationFunctions. A function is a rule that gives exactly one output number to each input number.
Functions A function is a rule that gives exactly one output number to each input number. Why it is important to us? The set of all input numbers to which the rule applies is called the domain of the function.
More informationLearning Objectives. c D. Poole and A. Mackworth 2010 Artificial Intelligence, Lecture 7.2, Page 1
Learning Objectives At the end of the class you should be able to: identify a supervised learning problem characterize how the prediction is a function of the error measure avoid mixing the training and
More informationMachine Learning. Yuh-Jye Lee. March 1, Lab of Data Science and Machine Intelligence Dept. of Applied Math. at NCTU
Machine Learning Yuh-Jye Lee Lab of Data Science and Machine Intelligence Dept. of Applied Math. at NCTU March 1, 2017 1 / 13 Bayes Rule Bayes Rule Assume that {B 1, B 2,..., B k } is a partition of S
More informationCheng Soon Ong & Christian Walder. Canberra February June 2018
Cheng Soon Ong & Christian Walder Research Group and College of Engineering and Computer Science Canberra February June 2018 (Many figures from C. M. Bishop, "Pattern Recognition and ") 1of 89 Part II
More information