Limits from l Hôpital rule: Shannon entropy as limit cases of Rényi and Tsallis entropies
|
|
- Letitia Ryan
- 5 years ago
- Views:
Transcription
1 Limits from l Hôpital rule: Shannon entropy as it cases of Rényi and Tsallis entropies Frank Nielsen École Polytechnique/Sony Computer Science Laboratorie, Inc August 6, 200 Let us prove that both Rényi extensive R α and Tsallis non-extensive T α entropies of order α tend to Shannon entropy when α using l Hôpital rule. Following the same principle, we then show that both Rényi and Tsallis information divergences tend to the celebrated Kullback-Leibler divergence when α. L Hôpital rule dates back to the 7th century, and states that the it of the indeterminate ratio of functions equals to the it of the ratio of their derivatives provided that (i) the its of both the numerator and denominator coincide (either to, 0, or ), and that (ii) the it of the ratio of the derivatives also exists. That is, if x α f(x) = x α g(x) = 0 and x α f (x)/g (x) = l exists, then f(x) x α g(x) = f (x) x α g (x) = l. (Although the its of f and g can alternatively be chosen as ± instead of 0, we shall use below the theorem to solve 0/0 indeterminate forms.) Rényi entropy Consider Rényi entropy [2] R α (P ) = log n α i= pα i (for α > 0 and α ), and let us prove that R α (P ) tends to Shannon entropy S(P ) = n i= p i log p i
2 when α. Set f(α) = log n i= pα i (for any fixed distribution P ) and g(α) = α. Then dg(α) = and dα df(α) dα = n i= d dα (pα i ) n i= pα i after applying the derivative chain rule. Since d dα (pα i ) = d dα eα log p i = (log p i )e α log p i = p α i log p i, we get n g (α) = p α n i log p i, and α g (α) = p i log p i. i= Since α f(α) = α g(α) = 0 and α g (α) = n i= p i log p i, we deduce from l Hôpital rule that α R α (P ) = H(P ). That is, Rényi entropy tends to Shannon entropy as α. 2 Tsallis entropy Consider now the Tsallis entropy [3] T α of order α (for α R)): T α (P ) = i= pα i α. Let us prove using l Hôpital rule that α T α (P ) = S(p) = i p log p, Shannon entropy. For a fixed distribution P, write T α(p ) = f(α) g(α) with g(α) = α and g (α) =. We have f(α) = i= pα i and = i pα i log p i since (p α i ) = (e α log p i ) = p α i log p i. Since both its of f(α) and g(α) tend to 0 as α, we can apply l Hôpital rule and get α T α (P ) = α g (α) = i p i log p i = H(P ), namely Shannon entropy. 3 Rényi and Tsallis information divergences Based on Rényi and Tsallis entropies, we define the corresponding information divergences [4, ]: i= R α (P : Q) = T α (P : Q) = α log i α ( i p α i q α i, α > 0 and α 0 p α i q α i ), α R. 2
3 Let us again apply l Hôpital rule to show that those parametric information divergences tend to the Kullback-Leibler divergence when α. Consider f(α) = i= pα i q α i (for fixed distributions P and Q), using the derivative chain rule we get = i pα i q α i log p i p α i q α i log = i pα i q α i log p i using the fact that (q α i ) = (e ( α) log ) = log (e ( α) log ) = q α i log. Let g(α) = α, and r(α) = log f(α). We have g (α) = and r (α) = f(α) = i pα i q α i log p i q i i= pα i q α i. We are ready to apply l Hôpital rule since α r(x) = α g(x) = 0 and α r (x) = i p i log p i and α g (x) = to conclude that Rényi information divergence tends to α R α(p : Q) = i p i log p i = KL(P : Q), i.e., the Kullback-Leibler divergence, also known as the relative entropy with respect to Shannon entropy. Similarly, for Tsallis information divergence, we set t(α) = f(α) = i pα i q α i and get t (α) = = i pα i q α i log p i. We apply l Hôpital rule and conclude that Tsallis information tends to the Kullback-Leibler divergence: α T α(p : Q) = i p i log p i = KL(P : Q). Note that those Rényi and Tsallis information divergences are mutually related to each other by monotonic transformations:, R α (P : Q) = log( + (α )T (P : Q)), () α T α (P : Q) = e(α )Rα(P :Q) (2) α In particular for α, using a first-order approximation log(+x) x 0 x, we find that R α(p : Q) α α α (α )T (P : Q) α T α (P : Q) (Similarly, we have exp(x) x + x, so that using E q.?? we deduce that T α (P : Q) α R α (P : Q).) 3
4 Although both extensive Rényi and non-extensive Tsallis information divergences meet in the it case α =, those divergences behave very differently from the classical Kullback-Leibler divergence. Indeed, the Kullback- Leibler divergence can be decomposed as KL(P : Q) = H(P, Q) H(P, P ), that is the cross-entropy H(P : Q) = i p i log minus the entropy H(P ) = H(P, P ) = i p i log p i. Such a sum-decomposition does not apply to Rényi/Tsallis information divergences. (Unfortunately, Rényi/Tsallis information divergences are also sometimes misleadingly called cross-entropies in the literature. Indeed, this is confusing since in that case the entropy cannot be considered as a self cross-entropy, that would otherwise yields an inconsistent zero!) 4 Proof of l Hôpital rule Consider the case of the indeterminate ratio 0/0. The proof of l Hôpital rule relies on Cauchy mean value theorem that states for a C -function f on [a, b] there exists c (a, b) such that f (c) = f(b) f(a). b a First, let us define by continuity f(α) = g(α) = 0, and let x α g (α) = l. There exists x (α δ, α + δ)\{α} such that both f (x) and g (x) exist and g (x) 0. Suppose without loss of generality that x (α, α + δ), then the mean value theorem implies that g(x) 0 (otherwise by contradiction, we would have y (α, x) with g (y) = 0). Thus according to the mean value theorem there exists a point ɛ x (α, x) such that f(x) g(x) = f (ɛ x ) g (ɛ x ) When x α, we have ɛ x α and since the ratio of the derivatives exists, we deduce that f(x) x α g(x) = f (ɛ x ) x α g (ɛ x ) = f (x) x α g (x) = l. 4
5 Figure : L Hospital textbook on infinitesimal calculus: Cover pages of the original anonymous edition of M.DC.XCVI=696 (left) and of the second edition of MDCCXVI=76 (right). 5 Historical notes Although the theorem is nowadays cited as l Hôpital rule, it is commonly believed to be partly his work since French nobleman Guillaume de l Hôpital (66-704, spelled l Hospital in old French) hired Swiss mathematician Johann Bernoulli ( ) to lecture him on mathematics. L Hôpital wrote and published in 696 the very first textbook on infinitesimal calculus (a precursor of modern differential calculus omitting voluntarily integration) titled: Analysis of the infinitely small to understand curves (see Figure ). The name of Bernoulli is acknowledged in the preface as the mathematician that pursued on the inspiring preinary work of Leibniz (646 76). 5
6 References [] Andrzej Cichocki, Anh Huy Phan, and Rafal Zdunek. Nonnegative Matrix and Tensor Factorizations: Applications to Exploratory Multi-way Data Analysis and Blind Source Separation; electronic version. Wiley, [2] Alfred Rényi. On measures of entropy and information. In Proc. 4th Berkeley Symp. Math. Stat. and Prob., volume, pages , 96. [3] Constantino Tsallis. Possible generalization of Boltzmann-Gibbs statistics. J. Stat. Physics, 52, 988. [4] Tim van Erven and Peter Harremoës. Rényi divergence and its properties. CoRR, abs/ ,
Convexity/Concavity of Renyi Entropy and α-mutual Information
Convexity/Concavity of Renyi Entropy and -Mutual Information Siu-Wai Ho Institute for Telecommunications Research University of South Australia Adelaide, SA 5095, Australia Email: siuwai.ho@unisa.edu.au
More informationMathematical Foundations of the Generalization of t-sne and SNE for Arbitrary Divergences
MACHINE LEARNING REPORTS Mathematical Foundations of the Generalization of t-sne and SNE for Arbitrary Report 02/2010 Submitted: 01.04.2010 Published:26.04.2010 T. Villmann and S. Haase University of Applied
More informationCalculus The Mean Value Theorem October 22, 2018
Calculus The Mean Value Theorem October, 018 Definitions Let c be a number in the domain D of a function f. Then f(c) is the (a) absolute maximum value of f on D, i.e. f(c) = max, if f(c) for all x in
More informationHistorical notes on calculus
Historical notes on calculus Dr. Vladimir Dotsenko Dr. Vladimir Dotsenko Historical notes on calculus 1 / 9 Descartes: Describing geometric figures by algebraic formulas 1637: René Descartes publishes
More informationMath 117: Honours Calculus I Fall, 2002 List of Theorems. a n k b k. k. Theorem 2.1 (Convergent Bounded) A convergent sequence is bounded.
Math 117: Honours Calculus I Fall, 2002 List of Theorems Theorem 1.1 (Binomial Theorem) For all n N, (a + b) n = n k=0 ( ) n a n k b k. k Theorem 2.1 (Convergent Bounded) A convergent sequence is bounded.
More information4.6. Indeterminate Forms and L Hôpital s Rule. 292 Chapter 4: Applications of Derivatives. Indeterminate Form 0/0
292 Chapter 4: Applications of Derivatives 4.6 Indeterminate Forms and L Hôpital s Rule HISTORICAL BIOGRAPHY Guillaume François Antoine de l Hôpital (66 74) John Bernoulli discovered a rule for calculating
More informationArkansas Tech University MATH 2924: Calculus II Dr. Marcel B. Finan
Arkansas Tech University MATH 2924: Calculus II Dr. Marcel B. Finan 8. Sequences We start this section by introducing the concept of a sequence and study its convergence. Convergence of Sequences. An infinite
More informationEntropy measures of physics via complexity
Entropy measures of physics via complexity Giorgio Kaniadakis and Flemming Topsøe Politecnico of Torino, Department of Physics and University of Copenhagen, Department of Mathematics 1 Introduction, Background
More informationMath 115 Spring 11 Written Homework 10 Solutions
Math 5 Spring Written Homework 0 Solutions. For following its, state what indeterminate form the its are in and evaluate the its. (a) 3x 4x 4 x x 8 Solution: This is in indeterminate form 0. Algebraically,
More informationMachine Learning. Lecture 02.2: Basics of Information Theory. Nevin L. Zhang
Machine Learning Lecture 02.2: Basics of Information Theory Nevin L. Zhang lzhang@cse.ust.hk Department of Computer Science and Engineering The Hong Kong University of Science and Technology Nevin L. Zhang
More informationLegendre transformation and information geometry
Legendre transformation and information geometry CIG-MEMO #2, v1 Frank Nielsen École Polytechnique Sony Computer Science Laboratorie, Inc http://www.informationgeometry.org September 2010 Abstract We explain
More informationMAT137 Calculus! Lecture 9
MAT137 Calculus! Lecture 9 Today we will study: Limits at infinity. L Hôpital s Rule. Mean Value Theorem. (11.5,11.6, 4.1) PS3 is due this Friday June 16. Next class: Applications of the Mean Value Theorem.
More information= 0. Theorem (Maximum Principle) If a holomorphic function f defined on a domain D C takes the maximum max z D f(z), then f is constant.
38 CHAPTER 3. HOLOMORPHIC FUNCTIONS Maximum Principle Maximum Principle was first proved for harmonic functions (i.e., the real part and the imaginary part of holomorphic functions) in Burkhardt s textbook
More informationAbout Closedness by Convolution of the Tsallis Maximizers
About Closedness by Convolution of the Tsallis Maximizers C. Vignat*, A. O. Hero III and J. A. Costa E.E.C.S., University of Michigan, Ann Arbor, USA (*) L.T.C.I., Université de Marne la Vallée, France
More informationOn Generalized Entropy Measures and Non-extensive Statistical Mechanics
First Prev Next Last On Generalized Entropy Measures and Non-extensive Statistical Mechanics A. M. MATHAI [Emeritus Professor of Mathematics and Statistics, McGill University, Canada, and Director, Centre
More informationChapter 8 Indeterminate Forms and Improper Integrals Math Class Notes
Chapter 8 Indeterminate Forms and Improper Integrals Math 1220-004 Class Notes Section 8.1: Indeterminate Forms of Type 0 0 Fact: The it of quotient is equal to the quotient of the its. (book page 68)
More informationCaculus 221. Possible questions for Exam II. March 19, 2002
Caculus 221 Possible questions for Exam II March 19, 2002 These notes cover the recent material in a style more like the lecture than the book. The proofs in the book are in section 1-11. At the end there
More informationShannon Entropy: Axiomatic Characterization and Application
Shannon Entropy: Axiomatic Characterization and Application C. G. Chakrabarti,Indranil Chakrabarty arxiv:quant-ph/0511171v1 17 Nov 2005 We have presented a new axiomatic derivation of Shannon Entropy for
More informationCalculus 1 Math 151 Week 10 Rob Rahm. Theorem 1.1. Rolle s Theorem. Let f be a function that satisfies the following three hypotheses:
Calculus 1 Math 151 Week 10 Rob Rahm 1 Mean Value Theorem Theorem 1.1. Rolle s Theorem. Let f be a function that satisfies the following three hypotheses: (1) f is continuous on [a, b]. (2) f is differentiable
More information5 Mutual Information and Channel Capacity
5 Mutual Information and Channel Capacity In Section 2, we have seen the use of a quantity called entropy to measure the amount of randomness in a random variable. In this section, we introduce several
More informationf(x)dx = lim f n (x)dx.
Chapter 3 Lebesgue Integration 3.1 Introduction The concept of integration as a technique that both acts as an inverse to the operation of differentiation and also computes areas under curves goes back
More informationIntroduction to Statistical Learning Theory
Introduction to Statistical Learning Theory In the last unit we looked at regularization - adding a w 2 penalty. We add a bias - we prefer classifiers with low norm. How to incorporate more complicated
More informationarxiv: v2 [math.ca] 8 Nov 2014
JOURNAL OF THE AMERICAN MATHEMATICAL SOCIETY Volume 00, Number 0, Pages 000 000 S 0894-0347(XX)0000-0 A NEW FRACTIONAL DERIVATIVE WITH CLASSICAL PROPERTIES arxiv:1410.6535v2 [math.ca] 8 Nov 2014 UDITA
More informationLog-concavity and the maximum entropy property of the Poisson distribution
Log-concavity and the maximum entropy property of the Poisson distribution arxiv:math/0603647v2 [math.pr] 11 Oct 2006 Oliver Johnson February 2, 2008 Abstract We prove that the Poisson distribution maximises
More informationBounds for entropy and divergence for distributions over a two-element set
Bounds for entropy and divergence for distributions over a two-element set Flemming Topsøe Department of Mathematics University of Copenhagen, Denmark 18.1.00 Abstract Three results dealing with probability
More informationChapter 8: Taylor s theorem and L Hospital s rule
Chapter 8: Taylor s theorem and L Hospital s rule Theorem: [Inverse Mapping Theorem] Suppose that a < b and f : [a, b] R. Given that f (x) > 0 for all x (a, b) then f 1 is differentiable on (f(a), f(b))
More informationISSN Article
Entropy 0, 3, 595-6; doi:0.3390/e3030595 OPEN ACCESS entropy ISSN 099-4300 www.mdpi.com/journal/entropy Article Entropy Measures vs. Kolmogorov Compleity Andreia Teieira,,, Armando Matos,3, André Souto,
More informationBeyond Newton and Leibniz: The Making of Modern Calculus. Anthony V. Piccolino, Ed. D. Palm Beach State College Palm Beach Gardens, Florida
Beyond Newton and Leibniz: The Making of Modern Calculus Anthony V. Piccolino, Ed. D. Palm Beach State College Palm Beach Gardens, Florida Calculus Before Newton & Leibniz Four Major Scientific Problems
More informationB553 Lecture 1: Calculus Review
B553 Lecture 1: Calculus Review Kris Hauser January 10, 2012 This course requires a familiarity with basic calculus, some multivariate calculus, linear algebra, and some basic notions of metric topology.
More informationLimits at Infinity. Horizontal Asymptotes. Definition (Limits at Infinity) Horizontal Asymptotes
Limits at Infinity If a function f has a domain that is unbounded, that is, one of the endpoints of its domain is ±, we can determine the long term behavior of the function using a it at infinity. Definition
More informationChapter 11: Sequences; Indeterminate Forms; Improper Integrals
Chapter 11: Sequences; Indeterminate Forms; Improper Integrals Section 11.1 The Least Upper Bound Axiom a. Least Upper Bound Axiom b. Examples c. Theorem 11.1.2 d. Example e. Greatest Lower Bound f. Theorem
More informationLeonhard Euler: Swiss man famous for mathematics, and not his chocolate
1 Jose Cabrera Dr. Shanyu Ji Math 4388 31 October 2016 Leonhard Euler: Swiss man famous for mathematics, and not his chocolate Leonhard Euler - one of the most revolutionary figures in 18th century mathematics.
More informationMATH 1231 MATHEMATICS 1B CALCULUS. Section 5: - Power Series and Taylor Series.
MATH 1231 MATHEMATICS 1B CALCULUS. Section 5: - Power Series and Taylor Series. The objective of this section is to become familiar with the theory and application of power series and Taylor series. By
More informationData Mining and Matrices
Data Mining and Matrices 6 Non-Negative Matrix Factorization Rainer Gemulla, Pauli Miettinen May 23, 23 Non-Negative Datasets Some datasets are intrinsically non-negative: Counters (e.g., no. occurrences
More informationSolution of the 8 th Homework
Solution of the 8 th Homework Sangchul Lee December 8, 2014 1 Preinary 1.1 A simple remark on continuity The following is a very simple and trivial observation. But still this saves a lot of words in actual
More informationPre-Calculus Notes from Week 6
1-105 Pre-Calculus Notes from Week 6 Logarithmic Functions: Let a > 0, a 1 be a given base (as in, base of an exponential function), and let x be any positive number. By our properties of exponential functions,
More informationProof that Fermat Prime Numbers are Infinite
Proof that Fermat Prime Numbers are Infinite Stephen Marshall 26 November 208 Abstract Fermat prime is a prime number that are a special case, given by the binomial number of the form: Fn = 2 2 n, for
More informationA Holevo-type bound for a Hilbert Schmidt distance measure
Journal of Quantum Information Science, 205, *,** Published Online **** 204 in SciRes. http://www.scirp.org/journal/**** http://dx.doi.org/0.4236/****.204.***** A Holevo-type bound for a Hilbert Schmidt
More informationExperimental Design to Maximize Information
Experimental Design to Maximize Information P. Sebastiani and H.P. Wynn Department of Mathematics and Statistics University of Massachusetts at Amherst, 01003 MA Department of Statistics, University of
More informationSolutions Final Exam May. 14, 2014
Solutions Final Exam May. 14, 2014 1. Determine whether the following statements are true or false. Justify your answer (i.e., prove the claim, derive a contradiction or give a counter-example). (a) (10
More informationLecture 3: Lower Bounds for Bandit Algorithms
CMSC 858G: Bandits, Experts and Games 09/19/16 Lecture 3: Lower Bounds for Bandit Algorithms Instructor: Alex Slivkins Scribed by: Soham De & Karthik A Sankararaman 1 Lower Bounds In this lecture (and
More informationLEBESGUE MEASURE AND L2 SPACE. Contents 1. Measure Spaces 1 2. Lebesgue Integration 2 3. L 2 Space 4 Acknowledgments 9 References 9
LBSGU MASUR AND L2 SPAC. ANNI WANG Abstract. This paper begins with an introduction to measure spaces and the Lebesgue theory of measure and integration. Several important theorems regarding the Lebesgue
More informationl Hǒpital s Rule and Limits of Riemann Sums (Textbook Supplement)
l Hǒpital s Rule and Limits of Riemann Sums Textbook Supplement The 0/0 Indeterminate Form and l Hǒpital s Rule Some weeks back, we already encountered a fundamental 0/0 indeterminate form, namely the
More informationEntropy and Ergodic Theory Lecture 4: Conditional entropy and mutual information
Entropy and Ergodic Theory Lecture 4: Conditional entropy and mutual information 1 Conditional entropy Let (Ω, F, P) be a probability space, let X be a RV taking values in some finite set A. In this lecture
More informationarxiv: v4 [cs.it] 17 Oct 2015
Upper Bounds on the Relative Entropy and Rényi Divergence as a Function of Total Variation Distance for Finite Alphabets Igal Sason Department of Electrical Engineering Technion Israel Institute of Technology
More informationCalculus I. When the following condition holds: if and only if
Calculus I I. Limits i) Notation: The limit of f of x, as x approaches a, is equal to L. ii) Formal Definition: Suppose f is defined on some open interval, which includes the number a. Then When the following
More information(x 3)(x + 5) = (x 3)(x 1) = x + 5. sin 2 x e ax bx 1 = 1 2. lim
SMT Calculus Test Solutions February, x + x 5 Compute x x x + Answer: Solution: Note that x + x 5 x x + x )x + 5) = x )x ) = x + 5 x x + 5 Then x x = + 5 = Compute all real values of b such that, for fx)
More informationLimit. Chapter Introduction
Chapter 9 Limit Limit is the foundation of calculus that it is so useful to understand more complicating chapters of calculus. Besides, Mathematics has black hole scenarios (dividing by zero, going to
More information6.1 Main properties of Shannon entropy. Let X be a random variable taking values x in some alphabet with probabilities.
Chapter 6 Quantum entropy There is a notion of entropy which quantifies the amount of uncertainty contained in an ensemble of Qbits. This is the von Neumann entropy that we introduce in this chapter. In
More informationCarefully tear this page off the rest of your exam. UNIVERSITY OF TORONTO SCARBOROUGH MATA37H3 : Calculus for Mathematical Sciences II
Carefully tear this page off the rest of your exam. UNIVERSITY OF TORONTO SCARBOROUGH MATA37H3 : Calculus for Mathematical Sciences II REFERENCE SHEET The (natural logarithm and exponential functions (
More informationarxiv:math/ v1 [math.st] 19 Jan 2005
ON A DIFFERENCE OF JENSEN INEQUALITY AND ITS APPLICATIONS TO MEAN DIVERGENCE MEASURES INDER JEET TANEJA arxiv:math/05030v [math.st] 9 Jan 005 Let Abstract. In this paper we have considered a difference
More informationMAS221 Analysis, Semester 1,
MAS221 Analysis, Semester 1, 2018-19 Sarah Whitehouse Contents About these notes 2 1 Numbers, inequalities, bounds and completeness 2 1.1 What is analysis?.......................... 2 1.2 Irrational numbers.........................
More informationL Hospital-Type Rules for Monotonicity: Include Them into Calculus Texts!
L Hospital-Type Rules for Monotonicity: Include Them into Calculus Texts! Iosif Pinelis Michigan Technological University Houghton, Michigan 49931 ipinelis@mtu.edu 1 Introduction Ratios are ubiquitous.
More informationd(x n, x) d(x n, x nk ) + d(x nk, x) where we chose any fixed k > N
Problem 1. Let f : A R R have the property that for every x A, there exists ɛ > 0 such that f(t) > ɛ if t (x ɛ, x + ɛ) A. If the set A is compact, prove there exists c > 0 such that f(x) > c for all x
More informationStatistical Machine Learning Lectures 4: Variational Bayes
1 / 29 Statistical Machine Learning Lectures 4: Variational Bayes Melih Kandemir Özyeğin University, İstanbul, Turkey 2 / 29 Synonyms Variational Bayes Variational Inference Variational Bayesian Inference
More informationShi Feng Sheng Danny Wong
Exhibit C A Proof of the Fermat s Last Theorem Shi Feng Sheng Danny Wong Abstract: Prior to the Diophantine geometry, number theory (or arithmetic) was to study the patterns of the numbers and elementary
More informationMath 161 (33) - Final exam
Name: Id #: Mat 161 (33) - Final exam Fall Quarter 2015 Wednesday December 9, 2015-10:30am to 12:30am Instructions: Prob. Points Score possible 1 25 2 25 3 25 4 25 TOTAL 75 (BEST 3) Read eac problem carefully.
More information1. Theorem. (Archimedean Property) Let x be any real number. There exists a positive integer n greater than x.
Advanced Calculus I, Dr. Block, Chapter 2 notes. Theorem. (Archimedean Property) Let x be any real number. There exists a positive integer n greater than x. 2. Definition. A sequence is a real-valued function
More informationMATH1190 CALCULUS 1 - NOTES AND AFTERNOTES
MATH90 CALCULUS - NOTES AND AFTERNOTES DR. JOSIP DERADO. Historical background Newton approach - from physics to calculus. Instantaneous velocity. Leibniz approach - from geometry to calculus Calculus
More informationSection 3.3 Product and Quotient rules 1 Lecture. Dr. Abdulla Eid. College of Science. MATHS 101: Calculus I
Section 3.3 Product and Quotient rules 1 Lecture College of Science MATHS 101: Calculus I (University of Bahrain) Differentiation Rules 1 / 12 Motivation Goal: We want to derive rules to find the derivative
More informationSection Signed Measures: The Hahn and Jordan Decompositions
17.2. Signed Measures 1 Section 17.2. Signed Measures: The Hahn and Jordan Decompositions Note. If measure space (X, M) admits measures µ 1 and µ 2, then for any α,β R where α 0,β 0, µ 3 = αµ 1 + βµ 2
More informationSemester University of Sheffield. School of Mathematics and Statistics
University of Sheffield School of Mathematics and Statistics MAS140: Mathematics (Chemical) MAS15: Civil Engineering Mathematics MAS15: Essential Mathematical Skills & Techniques MAS156: Mathematics (Electrical
More informationMathematics for Business and Economics - I. Chapter 5. Functions (Lecture 9)
Mathematics for Business and Economics - I Chapter 5. Functions (Lecture 9) Functions The idea of a function is this: a correspondence between two sets D and R such that to each element of the first set,
More informationFrom Calculus II: An infinite series is an expression of the form
MATH 3333 INTERMEDIATE ANALYSIS BLECHER NOTES 75 8. Infinite series of numbers From Calculus II: An infinite series is an expression of the form = a m + a m+ + a m+2 + ( ) Let us call this expression (*).
More informationMachine Learning Srihari. Information Theory. Sargur N. Srihari
Information Theory Sargur N. Srihari 1 Topics 1. Entropy as an Information Measure 1. Discrete variable definition Relationship to Code Length 2. Continuous Variable Differential Entropy 2. Maximum Entropy
More informationMAT137 Calculus! Lecture 20
official website http://uoft.me/mat137 MAT137 Calculus! Lecture 20 Today: 4.6 Concavity 4.7 Asypmtotes Next: 4.8 Curve Sketching Indeterminate Forms for Limits Which of the following are indeterminate
More informationIntroduction to Information Theory. Uncertainty. Entropy. Surprisal. Joint entropy. Conditional entropy. Mutual information.
L65 Dept. of Linguistics, Indiana University Fall 205 Information theory answers two fundamental questions in communication theory: What is the ultimate data compression? What is the transmission rate
More informationThe Information Bottleneck Revisited or How to Choose a Good Distortion Measure
The Information Bottleneck Revisited or How to Choose a Good Distortion Measure Peter Harremoës Centrum voor Wiskunde en Informatica PO 94079, 1090 GB Amsterdam The Nederlands PHarremoes@cwinl Naftali
More informationDept. of Linguistics, Indiana University Fall 2015
L645 Dept. of Linguistics, Indiana University Fall 2015 1 / 28 Information theory answers two fundamental questions in communication theory: What is the ultimate data compression? What is the transmission
More informationMath 221 Notes on Rolle s Theorem, The Mean Value Theorem, l Hôpital s rule, and the Taylor-Maclaurin formula. 1. Two theorems
Math 221 Notes on Rolle s Theorem, The Mean Value Theorem, l Hôpital s rule, and the Taylor-Maclaurin formula 1. Two theorems Rolle s Theorem. If a function y = f(x) is differentiable for a x b and if
More informationTight Bounds for Symmetric Divergence Measures and a Refined Bound for Lossless Source Coding
APPEARS IN THE IEEE TRANSACTIONS ON INFORMATION THEORY, FEBRUARY 015 1 Tight Bounds for Symmetric Divergence Measures and a Refined Bound for Lossless Source Coding Igal Sason Abstract Tight bounds for
More informationLiterature on Bregman divergences
Literature on Bregman divergences Lecture series at Univ. Hawai i at Mānoa Peter Harremoës February 26, 2016 Information divergence was introduced by Kullback and Leibler [25] and later Kullback started
More informationFoundations of Analysis. Joseph L. Taylor. University of Utah
Foundations of Analysis Joseph L. Taylor University of Utah Contents Preface vii Chapter 1. The Real Numbers 1 1.1. Sets and Functions 2 1.2. The Natural Numbers 8 1.3. Integers and Rational Numbers 16
More informationWasserstein GAN. Juho Lee. Jan 23, 2017
Wasserstein GAN Juho Lee Jan 23, 2017 Wasserstein GAN (WGAN) Arxiv submission Martin Arjovsky, Soumith Chintala, and Léon Bottou A new GAN model minimizing the Earth-Mover s distance (Wasserstein-1 distance)
More informationCS 591, Lecture 2 Data Analytics: Theory and Applications Boston University
CS 591, Lecture 2 Data Analytics: Theory and Applications Boston University Charalampos E. Tsourakakis January 25rd, 2017 Probability Theory The theory of probability is a system for making better guesses.
More informationESTIMATION OF ENTROPIES AND DIVERGENCES VIA NEAREST NEIGHBORS. 1. Introduction
Tatra Mt. Math. Publ. 39 (008), 65 73 t m Mathematical Publications ESTIMATION OF ENTROPIES AND DIVERGENCES VIA NEAREST NEIGHBORS Nikolai Leonenko Luc Pronzato Vippal Savani ABSTRACT. We extend the results
More informationHyperreals and a Brief Introduction to Non-Standard Analysis Math 336
Hyperreals and a Brief Introduction to Non-Standard Analysis Math 336 Gianni Krakoff June 8, 2015 Abstract The hyperreals are a number system extension of the real number system. With this number system
More informationMath 300 Introduction to Mathematical Reasoning Autumn 2017 Inverse Functions
Math 300 Introduction to Mathematical Reasoning Autumn 2017 Inverse Functions Please read this pdf in place of Section 6.5 in the text. The text uses the term inverse of a function and the notation f 1
More informationMT5836 Galois Theory MRQ
MT5836 Galois Theory MRQ May 3, 2017 Contents Introduction 3 Structure of the lecture course............................... 4 Recommended texts..................................... 4 1 Rings, Fields and
More informationJournal of Inequalities in Pure and Applied Mathematics
Journal of Inequalities in Pure and Applied Mathematics http://jipam.vu.edu.au/ Volume, Issue, Article 5, 00 BOUNDS FOR ENTROPY AND DIVERGENCE FOR DISTRIBUTIONS OVER A TWO-ELEMENT SET FLEMMING TOPSØE DEPARTMENT
More informationChapter 2: Entropy and Mutual Information. University of Illinois at Chicago ECE 534, Natasha Devroye
Chapter 2: Entropy and Mutual Information Chapter 2 outline Definitions Entropy Joint entropy, conditional entropy Relative entropy, mutual information Chain rules Jensen s inequality Log-sum inequality
More informationRolle s theorem: from a simple theorem to an extremely powerful tool
Rolle s theorem: from a simple theorem to an extremely powerful tool Christiane Rousseau, Université de Montréal November 10, 2011 1 Introduction How many solutions an equation or a system of equations
More informationExample 9 Algebraic Evaluation for Example 1
A Basic Principle Consider the it f(x) x a If you have a formula for the function f and direct substitution gives the indeterminate form 0, you may be able to evaluate the it algebraically. 0 Principle
More informationMATH 409 Advanced Calculus I Lecture 16: Mean value theorem. Taylor s formula.
MATH 409 Advanced Calculus I Lecture 16: Mean value theorem. Taylor s formula. Points of local extremum Let f : E R be a function defined on a set E R. Definition. We say that f attains a local maximum
More informationSection 4.2: The Mean Value Theorem
Section 4.2: The Mean Value Theorem Before we continue with the problem of describing graphs using calculus we shall briefly pause to examine some interesting applications of the derivative. In previous
More informationThe Mathematics of CT-Scans
The Mathematics of CT-Scans Tomography has become one of the most important applications of mathematics to the problems of keeping us alive. Modern medicine relies heavily on imaging methods, beginning
More informationarxiv:math-ph/ v1 30 May 2005
Nongeneralizability of Tsallis Entropy by means of Kolmogorov-Nagumo averages under pseudo-additivity arxiv:math-ph/0505078v1 30 May 2005 Ambedkar Dukkipati, 1 M. Narsimha Murty,,2 Shalabh Bhatnagar 3
More informationFunctional Limits and Continuity
Chapter 4 Functional Limits and Continuity 4.1 Discussion: Examples of Dirichlet and Thomae Although it is common practice in calculus courses to discuss continuity before differentiation, historically
More informationter. on Can we get a still better result? Yes, by making the rectangles still smaller. As we make the rectangles smaller and smaller, the
Area and Tangent Problem Calculus is motivated by two main problems. The first is the area problem. It is a well known result that the area of a rectangle with length l and width w is given by A = wl.
More informationMathematics 102 Fall 1999 The formal rules of calculus The three basic rules The sum rule. The product rule. The composition rule.
Mathematics 02 Fall 999 The formal rules of calculus So far we have calculated the derivative of each function we have looked at all over again from scratch, applying what is essentially the definition
More information4.3. Riemann Sums. Riemann Sums. Riemann Sums and Definite Integrals. Objectives
4.3 Riemann Sums and Definite Integrals Objectives Understand the definition of a Riemann sum. Evaluate a definite integral using limits & Riemann Sums. Evaluate a definite integral using geometric formulas
More informationSigned Measures. Chapter Basic Properties of Signed Measures. 4.2 Jordan and Hahn Decompositions
Chapter 4 Signed Measures Up until now our measures have always assumed values that were greater than or equal to 0. In this chapter we will extend our definition to allow for both positive negative values.
More informationL Hôpital s Rules and Taylor s Theorem for Product Calculus
Hôpital s Rules and Taylor s Theorem for Product Calculus Introduction Alex B Twist Michael Z Spivey University of Puget Sound Tacoma, Washington 9846 This paper is a continuation of the second author
More information11.6: Ratio and Root Tests Page 1. absolutely convergent, conditionally convergent, or divergent?
.6: Ratio and Root Tests Page Questions ( 3) n n 3 ( 3) n ( ) n 5 + n ( ) n e n ( ) n+ n2 2 n Example Show that ( ) n n ln n ( n 2 ) n + 2n 2 + converges for all x. Deduce that = 0 for all x. Solutions
More informationM2P1 Analysis II (2005) Dr M Ruzhansky List of definitions, statements and examples. Chapter 1: Limits and continuity.
M2P1 Analysis II (2005) Dr M Ruzhansky List of definitions, statements and examples. Chapter 1: Limits and continuity. This chapter is mostly the revision of Chapter 6 of M1P1. First we consider functions
More informationLemma 15.1 (Sign preservation Lemma). Suppose that f : E R is continuous at some a R.
15. Intermediate Value Theorem and Classification of discontinuities 15.1. Intermediate Value Theorem. Let us begin by recalling the definition of a function continuous at a point of its domain. Definition.
More informationThe Mean Value Inequality (without the Mean Value Theorem)
The Mean Value Inequality (without the Mean Value Theorem) Michael Müger September 13, 017 1 Introduction Abstract Once one has defined the notion of differentiability of a function of one variable, it
More informationMath 117: Honours Calculus I Fall, 2012 List of Theorems. a n k b k. k. Theorem 2.1 (Convergent Bounded): A convergent sequence is bounded.
Math 117: Honours Calculus I Fall, 2012 List of Theorems Theorem 1.1 (Binomial Theorem): For all n N, (a+b) n = n k=0 ( ) n a n k b k. k Theorem 2.1 (Convergent Bounded): A convergent sequence is bounded.
More informationIowa State University. Instructor: Alex Roitershtein Summer Homework #5. Solutions
Math 50 Iowa State University Introduction to Real Analysis Department of Mathematics Instructor: Alex Roitershtein Summer 205 Homework #5 Solutions. Let α and c be real numbers, c > 0, and f is defined
More informationLecture 2: What is Proof?
Lecture 2: What is Proof? Math 295 08/26/16 Webster Proof and Its History 8/2016 1 / 1 Evolution of Proof Proof, a relatively new idea Modern mathematics could not be supported at its foundation, nor construct
More information