Multicriteria Framework for Robust-Stochastic Formulations of Optimization under Uncertainty
|
|
- Amy Chloe Waters
- 6 years ago
- Views:
Transcription
1 Multicriteria Framework for Robust-Stochastic Formulations of Optimization under Uncertainty Alexander Engau Mathematical and Statistical Sciences University of Colorado Denver th Street, Suite 600 Denver, CO 80202, USA 2013 INFORMS ANNUAL MEETING MINNEAPOLIS CONVENTION CENTER OCTOBER 6-9, 2013 Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 1 of 1
2 Optimization Under Uncertainty (possible interpretations) Optimization problems or methods with elements of uncertainity 1. Stochastic optimization: random input to objective / constraints uses statistical inference to compute statistically optimal solutions assumes a probability distribution and optimizes in expectation chance-constraints, recourse, sampling, scenarios, simulation 2. Robust optimization: optimizes worst-case over uncertainty set 3. Parametric programming: solution stability / sensitivity analysis 4. Probabilistic or heuristic methods and randomized algorithms 5. Fuzzy optimization: imprecise quantification of set membership 6. Multicriteria optimization/decision-making: unknown preference Different concepts use different notions of feasibility or optimality! Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 2 of 1
3 Optimization Under Uncertainty (scope of this talk) Let u U be a continuous or discrete random variable or vector: minimize f (x, u) subject to x X(u) Two possibilities to handle uncertainty in the constraints: Consider the deterministic feasible set: X = u U X(u) Define probabilistic chance constraints: P(x X) 1 ɛ Two possibilities to handle uncertainty in the objective: Robust approach: optimize the worst-case { } minimize max f (x, u): u U subject to x X Stochastic approach: optimize the expectation minimize E U [f (x, u)] = w(u)f (x, u)du subject to x X u U Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 3 of 1
4 Multicriteria Approach to Optimization Under Uncertainty Formulate multiobjective deterministic equivalent program (DEP): { } minimize f (x, u): u U subject to x X Three-way classification based on underlying uncertainty set U finite and discrete: multiple-objective programming (MOP) optimization in R d (outcomes are finite-dimensional real vectors) infinitely discrete: infinitely-many-objective program (IMOP) optimization in l p for p (outcomes are real sequences) continuous: uncountably-many-objective program (UMOP) optimization in L p (outcomes are real-valued functions) Definition (Efficiency under Uncertainty) A point x X is said to be (weakly, properly) efficient under uncertainty if it is (weakly, properly) Pareto optimal for DEP. Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 4 of 1
5 Illustration of Efficiency Concepts for a Biobjective Program (BOP) Pareto Points prevent improvement Weak Pareto Points prevent strict improvement Proper Pareto Points prevent unbounded rates of substitution Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 5 of 1
6 IMOPs, UMOPs, and Efficiency Concepts at INFORMS 2013 SA38, Margaret Wiecek, Erin Doolittle, Hervé Kerivin: Preserving Efficiency in Multiobjective Programming under Uncertainty Observation that the addition of (possibly infinite numbers of) new objectives may not alter the Pareto efficient sets. The original objectives and the additional objectives are related through one or more algebraic operations. SA39, MA07, Dan Iancu, Nikos Trichakis Pareto Efficiency in Robust Optimization Observation that robust solutions are only weakly efficient. Suggestion of a direction-type method to test and achieve Pareto efficiency (similar to Benson s method). WA49, Pekka Korhonen (tomorrow morning - you can still make it!) Tutorial: Efficiency Concepts in MCDM, DEA, and Finance Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 6 of 1
7 Finite and Discrete Case: Multiobjective Programming Let I = {1, 2,..., m}, U = {u i : i I}, and f i (x) := f (x, u i ): minimize (f 1 (x), f 2 (x),..., f m (x)) subject to x X Definition (Pareto Optimality and Ideal / Utopia Points) A point x in the set of feasible decisions X is called weakly Pareto optimal: f i (x ) > f i (x) f j (x ) f j (x) Pareto optimal (Pareto 1896): f i (x ) > f i (x) f j (x ) < f j (x) properly Pareto optimal (Geoffrion 1968): there exists M < f i (x ) > f i (x) f i (x ) f i (x) < M(f j (x) f j (x )) A point z or z in the outcome space R m is called ideal point: z := (z1, z 2,..., z m) with zi = min {f i (x): x X} utopia point: z < z or z = z ɛ Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 7 of 1
8 Scalarizations from Compromise Programming (Zeleny 1973) Let w 0 be a vector of weights and z be an ideal or utopia point: ( minimize (w i(f i (x) z i )) p) 1/p subject to x X i I p = 1: weighted sum (Geoffrion 1968) minimize i I w if i (x) subject to x X p = : weighted Tchebycheff norm (Bowman 1976) minimize max {v i (f i (x) z i )} subject to x X i I augmented Tchebycheff norm (Atkins, Choo, Steuer 1983) minimize max{v i (f i (x) z i )} + α w if i (x) subject to x X i I i I Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 8 of 1
9 Characterizations of Pareto Optimality/Efficiency Compromise programming with nonnegative weights: Unique optimal solutions are efficient. Optimal solutions with w > 0 are efficient. Optimal solutions with w 0 are weakly efficient. Tchebycheff norm with nonnegative weights and utopia points: Solutions are weakly efficient if and only if they are optimal (for some suitable weight vector v 0 and a utopia point z). Weighted-sum with nonnegative or strictly positive weights: Optimal solutions with w > 0 are properly efficient. Sufficient conditions are also necessary if problem is convex. Augmented weighted Tchebycheff norm with positive weights: Solutions are properly efficient if and only if they are optimal (for suitable weights v > 0, small α > 0, and a utopia point z). Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 9 of 1
10 Infinite and Discrete Case: Infinitely-Many-Objective Programming Let I = N = {1, 2, 3,...}, U = {u i : i I}, and f i (x) := f (x, u i ): minimize (f 1 (x), f 2 (x), f 3 (x),...) subject to x X Outcomes and utopia points are elements in sequence space l p. Weighted sum becomes a weighted (infinite) series: minimize i=1 w if i (x) = lim k k i=1 w if i (x) subject to x X Tchebycheff (max) norm becomes supremum norm: minimize sup{v i (f i (x) z i )} subject to x X Compromise programming results for (weak) efficiency carry over. Weighted-series results follow from the Hahn-Banach Theorem. Results for proper efficiency do not hold under limiting behavior. Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 10 of 1
11 Counterexample: Strictly positive weights give improper solutions Let I = {0, 1, 2...}, X = [0, 1], f 0 (x) = 1 x, f i (x) = x i for i 1. All x [0, 1] are efficient (f 0 is decreasing, f i are increasing). All x can be generated using a weighted series (technical): w 0 = 1 + log(1 r) > 0 w i = r i /i > 0 for i 1 x(r) = 1 + log(1 r) r (1 + log(1 r))r x = 1 is not properly efficient. Consequence for Stochastic Optimization: Efficient solutions in expectation may still allow for unbounded rates of substitution. Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 11 of 1
12 Proper Efficiency for Countably Many Objectives Geoffrion: propose a slightly restricted definition of efficiency that (a) eliminates efficient points of a certain anomalous type; and (b) lends itself to more satisfactory characterization. Remark on improper points: because there is but a finite number of criteria, the marginal gain for some criterion can be arbitrarily large relative to each of the marginal losses in other criteria. New Definition of Proper Efficiency in the Sense of Geoffrion Above intention is maintained if for each i, there exist M i < f i (x ) > f i (x) f i (x ) f i (x) < M i (f j (x) f j (x )) If I is finite, M = max{m i : i I} recovers the original definition. If I is infinite, this new definition maintains the characterization of proper efficiency by weighted series and Tchebycheff norms. Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 12 of 1
13 Continuous Case: Uncountably-Many-Objective Programming Let U R d be a compact (closed and bounded) uncertainty set: ( ) minimize f (x, u): u U subject to x X Outcomes and utopia points are elements in function space L p. Weighted sum becomes a weighted integral: minimize w(u)f (x, u)du subject to x X u U (here w(u) is measurable Radon-Nikodym derivative / density) Tchebycheff norm becomes uniform (sup/max) norm: { } minimize max v(u)(f (x, u) z(u)) subject to x X u U (here v(u) and z(u) are weight and ideal or utopia functions) Compromise programming results for (weak) efficiency still hold! Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 13 of 1
14 Geoffrion Proper Efficiency for Uncountably Many Objectives Winkler (2004) considers a compact set T X and Y C(T ): minimize y(t) subject to y Y (analogous to UMOP with T = U, Y = {y(t) := f (x, t) = f (x, u)}) ȳ is efficient: ȳ(t) > y(t) ȳ(t 0 ) < y(t 0 ) properly efficient: ȳ(t) > y(t) ȳ(t) y(t) < δ(y(t 0 ) ȳ(t 0 )) All characterizations fail again! X = {x 1}, T = [0, 1], w = 1 { (x + 1) 2 y x (t) = min t 1, 1 } 2x x Then T y x(t)dt = 0 for all y Y but ȳ = 0 is not properly efficient. Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 14 of 1
15 Almost Proper Efficiency for Uncountably Many Objectives Solutions generated by weighting methods are proper efficient a.e. New Definition of Proper Efficiency Almost Everywhere Element ȳ is properly efficient a.e. if for every ɛ > 0, there is δ > 0 ȳ(t) > y(t) ȳ(t) y(t) < δ(y(t 0 ) ȳ(t 0 )) for all t T \ V with L(V ) < ɛ (here L is the Lebesgue measure). With the above definition, we could prove the following results: We could also give counterexamples for reversed implications. Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 15 of 1
16 Consequences and Discussion The theory and methods of multicriteria optimization provide a new framework for standard optimization problems under uncertainty. Weighted sums, series, or integrals for stochastic formulations. Tchebycheff and supremum norms for robust formulations. Multicriteria decision-making can help analyze such problems. Stochastic uncertainty corresponds to preferential uncertainty. Statistical inference corresponds to preference articulation. There are interesting new theoretical and practical implications. No characterization of proper efficiency in infinite dimensions. Stochastic and robust solutions may allow unbounded tradeoffs. Questions and comments are always welcome! Alexander Engau University of Colorado Denver Multicriteria Framework for Robust-Stochastic Optimization 16 of 1
A Note on Robustness of the Min-Max Solution to Multiobjective Linear Programs
A Note on Robustness of the Min-Max Solution to Multiobjective Linear Programs Erin K. Doolittle, Karyn Muir, and Margaret M. Wiecek Department of Mathematical Sciences Clemson University Clemson, SC January
More informationGenerating epsilon-efficient solutions in multiobjective programming
DEPARTMENT OF MATHEMATICAL SCIENCES Clemson University, South Carolina, USA Technical Report TR2005 10 EWb Generating epsilon-efficient solutions in multiobjective programming A. Engau and M. M. Wiecek
More informationLocal Approximation of the Efficient Frontier in Robust Design
Local Approximation of the Efficient Frontier in Robust Design Jinhuan Zhang, Graduate Assistant Department of Mechanical Engineering Clemson University Margaret M. Wiecek, Associate Professor Department
More informationMath 5593 Linear Programming Weeks 13/14
Math 5593 Linear Programming Weeks 13/14 11./12. Stochastic and Multiobjective Programming University of Colorado Denver, Fall 2013, Prof. Engau 1 Stochastic Programming 2 Multiobjective Programming 3
More informationWłodzimierz Ogryczak. Warsaw University of Technology, ICCE ON ROBUST SOLUTIONS TO MULTI-OBJECTIVE LINEAR PROGRAMS. Introduction. Abstract.
Włodzimierz Ogryczak Warsaw University of Technology, ICCE ON ROBUST SOLUTIONS TO MULTI-OBJECTIVE LINEAR PROGRAMS Abstract In multiple criteria linear programming (MOLP) any efficient solution can be found
More informationComputing Efficient Solutions of Nonconvex Multi-Objective Problems via Scalarization
Computing Efficient Solutions of Nonconvex Multi-Objective Problems via Scalarization REFAIL KASIMBEYLI Izmir University of Economics Department of Industrial Systems Engineering Sakarya Caddesi 156, 35330
More informationMixed-Integer Multiobjective Process Planning under Uncertainty
Ind. Eng. Chem. Res. 2002, 41, 4075-4084 4075 Mixed-Integer Multiobjective Process Planning under Uncertainty Hernán Rodera, Miguel J. Bagajewicz,* and Theodore B. Trafalis University of Oklahoma, 100
More informationTIES598 Nonlinear Multiobjective Optimization A priori and a posteriori methods spring 2017
TIES598 Nonlinear Multiobjective Optimization A priori and a posteriori methods spring 2017 Jussi Hakanen jussi.hakanen@jyu.fi Contents A priori methods A posteriori methods Some example methods Learning
More informationTrajectorial Martingales, Null Sets, Convergence and Integration
Trajectorial Martingales, Null Sets, Convergence and Integration Sebastian Ferrando, Department of Mathematics, Ryerson University, Toronto, Canada Alfredo Gonzalez and Sebastian Ferrando Trajectorial
More informationSome new results in post-pareto analysis
Some new results in post-pareto analysis H. Bonnel Université de la Nouvelle-Calédonie (ERIM) South Pacific Optimization Meeting: SPOM 13 Newcastle, February 9-12, 2013 Outline 1 Introduction 2 Post-Pareto
More informationRobust Solutions to Uncertain Multiobjective Programs
Clemson University TigerPrints All Dissertations Dissertations 5-2018 Robust Solutions to Uncertain Multiobjective Programs Garrett M. Dranichak Clemson University, gdranic@g.clemson.edu Follow this and
More informationIN many real-life situations we come across problems with
Algorithm for Interval Linear Programming Involving Interval Constraints Ibraheem Alolyan Abstract In real optimization, we always meet the criteria of useful outcomes increasing or expenses decreasing
More informationExamples of Dual Spaces from Measure Theory
Chapter 9 Examples of Dual Spaces from Measure Theory We have seen that L (, A, µ) is a Banach space for any measure space (, A, µ). We will extend that concept in the following section to identify an
More information1 Stochastic Dynamic Programming
1 Stochastic Dynamic Programming Formally, a stochastic dynamic program has the same components as a deterministic one; the only modification is to the state transition equation. When events in the future
More informationChapter 2 An Overview of Multiple Criteria Decision Aid
Chapter 2 An Overview of Multiple Criteria Decision Aid Abstract This chapter provides an overview of the multicriteria decision aid paradigm. The discussion covers the main features and concepts in the
More informationInfinite-Horizon Discounted Markov Decision Processes
Infinite-Horizon Discounted Markov Decision Processes Dan Zhang Leeds School of Business University of Colorado at Boulder Dan Zhang, Spring 2012 Infinite Horizon Discounted MDP 1 Outline The expected
More informationLebesgue Integration: A non-rigorous introduction. What is wrong with Riemann integration?
Lebesgue Integration: A non-rigorous introduction What is wrong with Riemann integration? xample. Let f(x) = { 0 for x Q 1 for x / Q. The upper integral is 1, while the lower integral is 0. Yet, the function
More informationSemidefinite and Second Order Cone Programming Seminar Fall 2012 Project: Robust Optimization and its Application of Robust Portfolio Optimization
Semidefinite and Second Order Cone Programming Seminar Fall 2012 Project: Robust Optimization and its Application of Robust Portfolio Optimization Instructor: Farid Alizadeh Author: Ai Kagawa 12/12/2012
More informationSOME HISTORY OF STOCHASTIC PROGRAMMING
SOME HISTORY OF STOCHASTIC PROGRAMMING Early 1950 s: in applications of Linear Programming unknown values of coefficients: demands, technological coefficients, yields, etc. QUOTATION Dantzig, Interfaces
More informationPareto Efficiency in Robust Optimization
Pareto Efficiency in Robust Optimization Dan Iancu Graduate School of Business Stanford University joint work with Nikolaos Trichakis (HBS) 1/26 Classical Robust Optimization Typical linear optimization
More informationOn Computing Highly Robust Efficient Solutions
On Computing Highly Robust Efficient Solutions Garrett M. Dranichak a,, Margaret M. Wiecek a a Department of Mathematical Sciences, Clemson University, Clemson, SC 29634, United States Abstract Due to
More informationCone characterizations of approximate solutions in real-vector optimization
DEPARTMENT OF MATHEMATICAL SCIENCES Clemson University, South Carolina, USA Technical Report TR2005 10 EW Cone characterizations of approximate solutions in real-vector optimization A. Engau and M. M.
More informationProduct metrics and boundedness
@ Applied General Topology c Universidad Politécnica de Valencia Volume 9, No. 1, 2008 pp. 133-142 Product metrics and boundedness Gerald Beer Abstract. This paper looks at some possible ways of equipping
More informationWeak convergence. Amsterdam, 13 November Leiden University. Limit theorems. Shota Gugushvili. Generalities. Criteria
Weak Leiden University Amsterdam, 13 November 2013 Outline 1 2 3 4 5 6 7 Definition Definition Let µ, µ 1, µ 2,... be probability measures on (R, B). It is said that µ n converges weakly to µ, and we then
More informationMultiple Objective Linear Programming in Supporting Forest Management
Multiple Objective Linear Programming in Supporting Forest Management Pekka Korhonen (December1998) International Institute for Applied Systems Analysis A-2361 Laxenburg, AUSTRIA and Helsinki School of
More informationORIGINS OF STOCHASTIC PROGRAMMING
ORIGINS OF STOCHASTIC PROGRAMMING Early 1950 s: in applications of Linear Programming unknown values of coefficients: demands, technological coefficients, yields, etc. QUOTATION Dantzig, Interfaces 20,1990
More informationLecture 1. Stochastic Optimization: Introduction. January 8, 2018
Lecture 1 Stochastic Optimization: Introduction January 8, 2018 Optimization Concerned with mininmization/maximization of mathematical functions Often subject to constraints Euler (1707-1783): Nothing
More informationStochastic Dynamic Programming. Jesus Fernandez-Villaverde University of Pennsylvania
Stochastic Dynamic Programming Jesus Fernande-Villaverde University of Pennsylvania 1 Introducing Uncertainty in Dynamic Programming Stochastic dynamic programming presents a very exible framework to handle
More informationA Geometric Characterization of the Power of Finite Adaptability in Multistage Stochastic and Adaptive Optimization
MATHEMATICS OF OPERATIONS RESEARCH Vol. 36, No., February 20, pp. 24 54 issn 0364-765X eissn 526-547 360 0024 informs doi 0.287/moor.0.0482 20 INFORMS A Geometric Characterization of the Power of Finite
More informationFive Mini-Courses on Analysis
Christopher Heil Five Mini-Courses on Analysis Metrics, Norms, Inner Products, and Topology Lebesgue Measure and Integral Operator Theory and Functional Analysis Borel and Radon Measures Topological Vector
More informationMultiobjective Mixed-Integer Stackelberg Games
Solving the Multiobjective Mixed-Integer SCOTT DENEGRE TED RALPHS ISE Department COR@L Lab Lehigh University tkralphs@lehigh.edu EURO XXI, Reykjavic, Iceland July 3, 2006 Outline Solving the 1 General
More informationInternational Journal of Information Technology & Decision Making c World Scientific Publishing Company
International Journal of Information Technology & Decision Making c World Scientific Publishing Company A MIN-MAX GOAL PROGRAMMING APPROACH TO PRIORITY DERIVATION IN AHP WITH INTERVAL JUDGEMENTS DIMITRIS
More informationA New Fuzzy Positive and Negative Ideal Solution for Fuzzy TOPSIS
A New Fuzzy Positive and Negative Ideal Solution for Fuzzy TOPSIS MEHDI AMIRI-AREF, NIKBAKHSH JAVADIAN, MOHAMMAD KAZEMI Department of Industrial Engineering Mazandaran University of Science & Technology
More informationProblem set 1, Real Analysis I, Spring, 2015.
Problem set 1, Real Analysis I, Spring, 015. (1) Let f n : D R be a sequence of functions with domain D R n. Recall that f n f uniformly if and only if for all ɛ > 0, there is an N = N(ɛ) so that if n
More informationMultiobjective optimization methods
Multiobjective optimization methods Jussi Hakanen Post-doctoral researcher jussi.hakanen@jyu.fi spring 2014 TIES483 Nonlinear optimization No-preference methods DM not available (e.g. online optimization)
More informationApproximations for Pareto and Proper Pareto solutions and their KKT conditions
Approximations for Pareto and Proper Pareto solutions and their KKT conditions P. Kesarwani, P.K.Shukla, J. Dutta and K. Deb October 6, 2018 Abstract There has been numerous amount of studies on proper
More informationMATHS 730 FC Lecture Notes March 5, Introduction
1 INTRODUCTION MATHS 730 FC Lecture Notes March 5, 2014 1 Introduction Definition. If A, B are sets and there exists a bijection A B, they have the same cardinality, which we write as A, #A. If there exists
More informationOn the Power of Robust Solutions in Two-Stage Stochastic and Adaptive Optimization Problems
MATHEMATICS OF OPERATIONS RESEARCH Vol. 35, No., May 010, pp. 84 305 issn 0364-765X eissn 156-5471 10 350 084 informs doi 10.187/moor.1090.0440 010 INFORMS On the Power of Robust Solutions in Two-Stage
More informationExamples. 1.1 Cones in Vector Spaces
Examples. Cones in Vector Spaces Vector optimization in partially ordered spaces requires, among other things, that one studies properties of cones; as arguments recall: Conic approximation of sets, e.g.,
More informationOn Kusuoka Representation of Law Invariant Risk Measures
MATHEMATICS OF OPERATIONS RESEARCH Vol. 38, No. 1, February 213, pp. 142 152 ISSN 364-765X (print) ISSN 1526-5471 (online) http://dx.doi.org/1.1287/moor.112.563 213 INFORMS On Kusuoka Representation of
More informationTHE REFERENCE POINT METHOD WITH LEXICOGRAPHIC MIN-ORDERING OF INDIVIDUAL ACHIEVEMENTS
Włodzimierz Ogryczak THE REFERENCE POINT METHOD WITH LEXICOGRAPHIC MIN-ORDERING OF INDIVIDUAL ACHIEVEMENTS INTRODUCTION Typical multiple criteria optimization methods aggregate the individual outcomes
More informationCHAPTER V DUAL SPACES
CHAPTER V DUAL SPACES DEFINITION Let (X, T ) be a (real) locally convex topological vector space. By the dual space X, or (X, T ), of X we mean the set of all continuous linear functionals on X. By the
More informationMeasure Theory. John K. Hunter. Department of Mathematics, University of California at Davis
Measure Theory John K. Hunter Department of Mathematics, University of California at Davis Abstract. These are some brief notes on measure theory, concentrating on Lebesgue measure on R n. Some missing
More informationReal Analysis, 2nd Edition, G.B.Folland Elements of Functional Analysis
Real Analysis, 2nd Edition, G.B.Folland Chapter 5 Elements of Functional Analysis Yung-Hsiang Huang 5.1 Normed Vector Spaces 1. Note for any x, y X and a, b K, x+y x + y and by ax b y x + b a x. 2. It
More informationTheoretical Foundation of Uncertain Dominance
Theoretical Foundation of Uncertain Dominance Yang Zuo, Xiaoyu Ji 2 Uncertainty Theory Laboratory, Department of Mathematical Sciences Tsinghua University, Beijing 84, China 2 School of Business, Renmin
More informationReasoning with Uncertainty
Reasoning with Uncertainty Representing Uncertainty Manfred Huber 2005 1 Reasoning with Uncertainty The goal of reasoning is usually to: Determine the state of the world Determine what actions to take
More informationSelected Examples of CONIC DUALITY AT WORK Robust Linear Optimization Synthesis of Linear Controllers Matrix Cube Theorem A.
. Selected Examples of CONIC DUALITY AT WORK Robust Linear Optimization Synthesis of Linear Controllers Matrix Cube Theorem A. Nemirovski Arkadi.Nemirovski@isye.gatech.edu Linear Optimization Problem,
More informationFUNCTIONAL ANALYSIS LECTURE NOTES: WEAK AND WEAK* CONVERGENCE
FUNCTIONAL ANALYSIS LECTURE NOTES: WEAK AND WEAK* CONVERGENCE CHRISTOPHER HEIL 1. Weak and Weak* Convergence of Vectors Definition 1.1. Let X be a normed linear space, and let x n, x X. a. We say that
More informationCOHERENT APPROACHES TO RISK IN OPTIMIZATION UNDER UNCERTAINTY
COHERENT APPROACHES TO RISK IN OPTIMIZATION UNDER UNCERTAINTY Terry Rockafellar University of Washington, Seattle University of Florida, Gainesville Goal: a coordinated view of recent ideas in risk modeling
More informationStochastic dominance with imprecise information
Stochastic dominance with imprecise information Ignacio Montes, Enrique Miranda, Susana Montes University of Oviedo, Dep. of Statistics and Operations Research. Abstract Stochastic dominance, which is
More informationOn maxitive integration
On maxitive integration Marco E. G. V. Cattaneo Department of Mathematics, University of Hull m.cattaneo@hull.ac.uk Abstract A functional is said to be maxitive if it commutes with the (pointwise supremum
More informationNotions such as convergent sequence and Cauchy sequence make sense for any metric space. Convergent Sequences are Cauchy
Banach Spaces These notes provide an introduction to Banach spaces, which are complete normed vector spaces. For the purposes of these notes, all vector spaces are assumed to be over the real numbers.
More information3 (Due ). Let A X consist of points (x, y) such that either x or y is a rational number. Is A measurable? What is its Lebesgue measure?
MA 645-4A (Real Analysis), Dr. Chernov Homework assignment 1 (Due ). Show that the open disk x 2 + y 2 < 1 is a countable union of planar elementary sets. Show that the closed disk x 2 + y 2 1 is a countable
More informationA New Approach for Uncertain Multiobjective Programming Problem Based on P E Principle
INFORMATION Volume xx, Number xx, pp.54-63 ISSN 1343-45 c 21x International Information Institute A New Approach for Uncertain Multiobjective Programming Problem Based on P E Principle Zutong Wang 1, Jiansheng
More information5 Measure theory II. (or. lim. Prove the proposition. 5. For fixed F A and φ M define the restriction of φ on F by writing.
5 Measure theory II 1. Charges (signed measures). Let (Ω, A) be a σ -algebra. A map φ: A R is called a charge, (or signed measure or σ -additive set function) if φ = φ(a j ) (5.1) A j for any disjoint
More informationProblem Set 2: Solutions Math 201A: Fall 2016
Problem Set 2: s Math 201A: Fall 2016 Problem 1. (a) Prove that a closed subset of a complete metric space is complete. (b) Prove that a closed subset of a compact metric space is compact. (c) Prove that
More informationApproaches to Sensitivity Analysis in MOLP
I.J. Information echnology and Computer Science, 204, 03, 54-60 Published Online February 204 in MECS (http://www.mecs-press.org/) DOI: 0.585/ijitcs.204.03.07 Approaches to Sensitivity Analysis in MOLP
More information2D Decision-Making for Multi-Criteria Design Optimization
DEPARTMENT OF MATHEMATICAL SCIENCES Clemson University, South Carolina, USA Technical Report TR2006 05 EW 2D Decision-Making for Multi-Criteria Design Optimization A. Engau and M. M. Wiecek May 2006 This
More informationL p Spaces and Convexity
L p Spaces and Convexity These notes largely follow the treatments in Royden, Real Analysis, and Rudin, Real & Complex Analysis. 1. Convex functions Let I R be an interval. For I open, we say a function
More informationA DEA- COMPROMISE PROGRAMMING MODEL FOR COMPREHENSIVE RANKING
Journal of the Operations Research Society of Japan 2004, Vol. 47, No. 2, 73-81 2004 The Operations Research Society of Japan A DEA- COMPROMISE PROGRAMMING MODEL FOR COMPREHENSIVE RANKING Akihiro Hashimoto
More informationSequential Pareto Subdifferential Sum Rule And Sequential Effi ciency
Applied Mathematics E-Notes, 16(2016), 133-143 c ISSN 1607-2510 Available free at mirror sites of http://www.math.nthu.edu.tw/ amen/ Sequential Pareto Subdifferential Sum Rule And Sequential Effi ciency
More informationNew Reference-Neighbourhood Scalarization Problem for Multiobjective Integer Programming
BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 3 No Sofia 3 Print ISSN: 3-97; Online ISSN: 34-48 DOI:.478/cait-3- New Reference-Neighbourhood Scalariation Problem for Multiobjective
More informationAnalysis Comprehensive Exam Questions Fall 2008
Analysis Comprehensive xam Questions Fall 28. (a) Let R be measurable with finite Lebesgue measure. Suppose that {f n } n N is a bounded sequence in L 2 () and there exists a function f such that f n (x)
More informationProbability and Measure
Part II Year 2018 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2018 84 Paper 4, Section II 26J Let (X, A) be a measurable space. Let T : X X be a measurable map, and µ a probability
More informationReproducing Kernel Hilbert Spaces
9.520: Statistical Learning Theory and Applications February 10th, 2010 Reproducing Kernel Hilbert Spaces Lecturer: Lorenzo Rosasco Scribe: Greg Durrett 1 Introduction In the previous two lectures, we
More information+ 2x sin x. f(b i ) f(a i ) < ɛ. i=1. i=1
Appendix To understand weak derivatives and distributional derivatives in the simplest context of functions of a single variable, we describe without proof some results from real analysis (see [7] and
More informationPart II: Markov Processes. Prakash Panangaden McGill University
Part II: Markov Processes Prakash Panangaden McGill University 1 How do we define random processes on continuous state spaces? How do we define conditional probabilities on continuous state spaces? How
More informationAn introduction to some aspects of functional analysis
An introduction to some aspects of functional analysis Stephen Semmes Rice University Abstract These informal notes deal with some very basic objects in functional analysis, including norms and seminorms
More informationRandom Process Lecture 1. Fundamentals of Probability
Random Process Lecture 1. Fundamentals of Probability Husheng Li Min Kao Department of Electrical Engineering and Computer Science University of Tennessee, Knoxville Spring, 2016 1/43 Outline 2/43 1 Syllabus
More informationStability of optimization problems with stochastic dominance constraints
Stability of optimization problems with stochastic dominance constraints D. Dentcheva and W. Römisch Stevens Institute of Technology, Hoboken Humboldt-University Berlin www.math.hu-berlin.de/~romisch SIAM
More informationSet-based Min-max and Min-min Robustness for Multi-objective Robust Optimization
Proceedings of the 2017 Industrial and Systems Engineering Research Conference K. Coperich, E. Cudney, H. Nembhard, eds. Set-based Min-max and Min-min Robustness for Multi-objective Robust Optimization
More informationInformation Structures Preserved Under Nonlinear Time-Varying Feedback
Information Structures Preserved Under Nonlinear Time-Varying Feedback Michael Rotkowitz Electrical Engineering Royal Institute of Technology (KTH) SE-100 44 Stockholm, Sweden Email: michael.rotkowitz@ee.kth.se
More informationAdvanced Analysis Qualifying Examination Department of Mathematics and Statistics University of Massachusetts. Tuesday, January 16th, 2018
NAME: Advanced Analysis Qualifying Examination Department of Mathematics and Statistics University of Massachusetts Tuesday, January 16th, 2018 Instructions 1. This exam consists of eight (8) problems
More informationChapter 8. General Countably Additive Set Functions. 8.1 Hahn Decomposition Theorem
Chapter 8 General Countably dditive Set Functions In Theorem 5.2.2 the reader saw that if f : X R is integrable on the measure space (X,, µ) then we can define a countably additive set function ν on by
More informationInfinite-Horizon Average Reward Markov Decision Processes
Infinite-Horizon Average Reward Markov Decision Processes Dan Zhang Leeds School of Business University of Colorado at Boulder Dan Zhang, Spring 2012 Infinite Horizon Average Reward MDP 1 Outline The average
More informationDecomposability and time consistency of risk averse multistage programs
Decomposability and time consistency of risk averse multistage programs arxiv:1806.01497v1 [math.oc] 5 Jun 2018 A. Shapiro School of Industrial and Systems Engineering Georgia Institute of Technology Atlanta,
More informationHomework If the inverse T 1 of a closed linear operator exists, show that T 1 is a closed linear operator.
Homework 3 1 If the inverse T 1 of a closed linear operator exists, show that T 1 is a closed linear operator Solution: Assuming that the inverse of T were defined, then we will have to have that D(T 1
More informationMathematical Methods for Physics and Engineering
Mathematical Methods for Physics and Engineering Lecture notes for PDEs Sergei V. Shabanov Department of Mathematics, University of Florida, Gainesville, FL 32611 USA CHAPTER 1 The integration theory
More information1. Introduction. Consider the deterministic multi-objective linear semi-infinite program of the form (P ) V-min c (1.1)
ROBUST SOLUTIONS OF MULTI-OBJECTIVE LINEAR SEMI-INFINITE PROGRAMS UNDER CONSTRAINT DATA UNCERTAINTY M.A. GOBERNA, V. JEYAKUMAR, G. LI, AND J. VICENTE-PÉREZ Abstract. The multi-objective optimization model
More information2 (Bonus). Let A X consist of points (x, y) such that either x or y is a rational number. Is A measurable? What is its Lebesgue measure?
MA 645-4A (Real Analysis), Dr. Chernov Homework assignment 1 (Due 9/5). Prove that every countable set A is measurable and µ(a) = 0. 2 (Bonus). Let A consist of points (x, y) such that either x or y is
More informationStability theory is a fundamental topic in mathematics and engineering, that include every
Stability Theory Stability theory is a fundamental topic in mathematics and engineering, that include every branches of control theory. For a control system, the least requirement is that the system is
More informationOn deterministic reformulations of distributionally robust joint chance constrained optimization problems
On deterministic reformulations of distributionally robust joint chance constrained optimization problems Weijun Xie and Shabbir Ahmed School of Industrial & Systems Engineering Georgia Institute of Technology,
More informationHilbert Space Methods Used in a First Course in Quantum Mechanics A Recap WHY ARE WE HERE? QUOTE FROM WIKIPEDIA
Hilbert Space Methods Used in a First Course in Quantum Mechanics A Recap Larry Susanka Table of Contents Why Are We Here? The Main Vector Spaces Notions of Convergence Topological Vector Spaces Banach
More informationBest Guaranteed Result Principle and Decision Making in Operations with Stochastic Factors and Uncertainty
Stochastics and uncertainty underlie all the processes of the Universe. N.N.Moiseev Best Guaranteed Result Principle and Decision Making in Operations with Stochastic Factors and Uncertainty by Iouldouz
More informationTHE RADON NIKODYM PROPERTY AND LIPSCHITZ EMBEDDINGS
THE RADON NIKODYM PROPERTY AND LIPSCHITZ EMBEDDINGS Motivation The idea here is simple. Suppose we have a Lipschitz homeomorphism f : X Y where X and Y are Banach spaces, namely c 1 x y f (x) f (y) c 2
More informationDistributionally Robust Discrete Optimization with Entropic Value-at-Risk
Distributionally Robust Discrete Optimization with Entropic Value-at-Risk Daniel Zhuoyu Long Department of SEEM, The Chinese University of Hong Kong, zylong@se.cuhk.edu.hk Jin Qi NUS Business School, National
More information3. (a) What is a simple function? What is an integrable function? How is f dµ defined? Define it first
Math 632/6321: Theory of Functions of a Real Variable Sample Preinary Exam Questions 1. Let (, M, µ) be a measure space. (a) Prove that if µ() < and if 1 p < q
More informationVector Spaces. Vector space, ν, over the field of complex numbers, C, is a set of elements a, b,..., satisfying the following axioms.
Vector Spaces Vector space, ν, over the field of complex numbers, C, is a set of elements a, b,..., satisfying the following axioms. For each two vectors a, b ν there exists a summation procedure: a +
More informationRobust Solutions to Multi-Objective Linear Programs with Uncertain Data
Robust Solutions to Multi-Objective Linear Programs with Uncertain Data M.A. Goberna yz V. Jeyakumar x G. Li x J. Vicente-Pérez x Revised Version: October 1, 2014 Abstract In this paper we examine multi-objective
More informationHausdorff Measure. Jimmy Briggs and Tim Tyree. December 3, 2016
Hausdorff Measure Jimmy Briggs and Tim Tyree December 3, 2016 1 1 Introduction In this report, we explore the the measurement of arbitrary subsets of the metric space (X, ρ), a topological space X along
More informationConstruction of a general measure structure
Chapter 4 Construction of a general measure structure We turn to the development of general measure theory. The ingredients are a set describing the universe of points, a class of measurable subsets along
More informationFunctional Analysis. Martin Brokate. 1 Normed Spaces 2. 2 Hilbert Spaces The Principle of Uniform Boundedness 32
Functional Analysis Martin Brokate Contents 1 Normed Spaces 2 2 Hilbert Spaces 2 3 The Principle of Uniform Boundedness 32 4 Extension, Reflexivity, Separation 37 5 Compact subsets of C and L p 46 6 Weak
More informationOn the Power of Robust Solutions in Two-Stage Stochastic and Adaptive Optimization Problems
MATHEMATICS OF OPERATIONS RESEARCH Vol. xx, No. x, Xxxxxxx 00x, pp. xxx xxx ISSN 0364-765X EISSN 156-5471 0x xx0x 0xxx informs DOI 10.187/moor.xxxx.xxxx c 00x INFORMS On the Power of Robust Solutions in
More informationCHAPTER 6. Differentiation
CHPTER 6 Differentiation The generalization from elementary calculus of differentiation in measure theory is less obvious than that of integration, and the methods of treating it are somewhat involved.
More informationMulticriteria Decision Making Based on Fuzzy Relations
BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 8, No 4 Sofia 2008 Multicriteria Decision Maing Based on Fuzzy Relations Vania Peneva, Ivan Popchev Institute of Information
More informationSome Properties of the Augmented Lagrangian in Cone Constrained Optimization
MATHEMATICS OF OPERATIONS RESEARCH Vol. 29, No. 3, August 2004, pp. 479 491 issn 0364-765X eissn 1526-5471 04 2903 0479 informs doi 10.1287/moor.1040.0103 2004 INFORMS Some Properties of the Augmented
More informationSigned Measures and Complex Measures
Chapter 8 Signed Measures Complex Measures As opposed to the measures we have considered so far, a signed measure is allowed to take on both positive negative values. To be precise, if M is a σ-algebra
More informationA Bicriteria Approach to Robust Optimization
A Bicriteria Approach to Robust Optimization André Chassein and Marc Goerigk Fachbereich Mathematik, Technische Universität Kaiserslautern, Germany Abstract The classic approach in robust optimization
More informationUTILITY OPTIMIZATION IN A FINITE SCENARIO SETTING
UTILITY OPTIMIZATION IN A FINITE SCENARIO SETTING J. TEICHMANN Abstract. We introduce the main concepts of duality theory for utility optimization in a setting of finitely many economic scenarios. 1. Utility
More informationS. DUTTA AND T. S. S. R. K. RAO
ON WEAK -EXTREME POINTS IN BANACH SPACES S. DUTTA AND T. S. S. R. K. RAO Abstract. We study the extreme points of the unit ball of a Banach space that remain extreme when considered, under canonical embedding,
More information