1 The Primal and Dual of an Optimization Problem

Size: px
Start display at page:

Download "1 The Primal and Dual of an Optimization Problem"

Transcription

1 CS 189 Itroductio to Machie Learig Fall 2017 Note 18 Previously, i our ivestigatio of SVMs, we forulated a costraied optiizatio proble that we ca solve to fid the optial paraeters for our hyperplae decisio boudary. Recall the setup of soft-argi SVMs: y i s: ±1, represetig positive or egative class x i s: feature vectors i R d ξ i s: slack variables represetig how uch a x i is allowed to violate the argi C: a hyperparaeter describig how severely we pealize slack The optiizatio proble for w R d ad t R, the paraeters of the SVM: 1 i w,t,ξ i 2 w 2 +C ξ i s.t. y i (w T x i +t) 1 ξ i i ξ i 0 Now, we will ivestigate the dual of this proble, which will otivate our discussio of kerels. Before we do so, we first have to uderstad the prial ad dual of a optiizatio proble. i 1 The Prial ad Dual of a Optiizatio Proble All optiizatio probles ad be expressed i the stadard for i x f 0 (x) s.t. f i (x) 0 i = 1,..., h j (x) = 0 j = 1,..., (1) For the purposes of our discussio, assue that x R d. The ai copoets of a optiizatio proble are: The objective fuctio f 0 (x) The iequality costraits: expressios ivolvig f i (x) The equality costraits: expressios ivolvig h j (x) CS 189, Fall 2017, Note 18 1

2 Workig with the costraits ca be cubersoe ad challegig to aipulate, ad it would be ideal if we could soehow tur this costraied optiizatio proble ito a ucostraied oe. Oe idea is to re-express the optiizatio proble ito i x L (x) (2) where L (x) = { f 0 (x) if f i (x) 0, i [1,] ad h j (x) = 0, j [1,] otherwise Note that the ucostraied optiizatio proble above is equivalet to the origial costraied proble. Eve though the ucostraied proble cosiders values that violate the costraits (ad therefore are ot i the feasible set for the costraied optiizatio proble), it will effectively igore the because they are treated as i a iiizatio proble. Eve though we are ow dealig with a ucostraied proble, it still is difficult to solve the optiizatio proble, because we still have to deal with all of the casework i the objective fuctio L (x). I order to solve this issue, with have to itroduce dual variables, specifically oe set of dual variables for the equality costraits, ad oe set for the iequality costraits. First, let s deal with the equality costraits. If we oly take ito accout the dual variables for the equality costraits, the optiizatio proble ow becoes i x axl (x,ν) (3) ν where L (x,ν) = { f 0 (x) + ν jh j (x) if f i (x) 0, i [1,] otherwise We are still workig with a ucostraied optiizatio proble, except that ow, we are optiizig over two sets of variables: the prial variables x R d ad the dual variables ν R. Also ote that the optiizatio proble has ow becoe a ested oe, a ier optiizatio proble the axiizes over the dual variables, ad a outer optiizatio proble that iiizes over the prial variables. Let s exaie why this optiizatio proble is equivalet to the origial costraied optiizatio proble: Ay x that violates the iequality costraits is still treated as by the outer iiizatio proble over x ad therefore igored For ay x that violates the equality costraits (eaig that j s.t. h j (x) 0), the ier axiizatio proble over ν ca choose ν j as if h j (x) > 0 (or ν j as if h j (x) < 0) to cause the ier axiizatio blow off to, therefore beig igored by the outer iiizatio over x For ay x that does ot violate ay of the equality or iequality costraits, the ier axiizatio proble over ν is siply equal to f 0 (x) CS 189, Fall 2017, Note 18 2

3 This solutio coes at a cost i a effort to reove the equality costraits, we had to add i dual variables, oe for each iequality costrait. With this i id, let s try to do the sae for the iequality costraits. Addig i dual variable λ i to represet each iequality costrait, we ow have iax x λ,ν L (x,λ,ν) = f 0 (x) + s.t. λ i 0 i = 1,..., λ i f i (x) + ν j h j (x) For coveiece, we ca place the costraits ivolvig λ ito the optiizatio variable. i x ax L (x,λ,ν) = f 0(x) + λ 0,ν λ i f i (x) + ν j h j (x) This optiizatio proble above is otherwise kow as the prial, ad its optial value is ideed equivalet to that of the origial costraied optiizatio proble. p = i x We ca verify that this is ideed the case: ax L (x,λ,ν) λ 0,ν For ay x that violates the iequality costraits (eaig that i [1,] s.t. f i (x) > 0), the ier axiizatio proble over λ ca choose λ i as to cause the ier axiizatio blow off to, therefore beig igored by the outer iiizatio over x For ay x that violates the equality costraits (eaig that j s.t. h j (x) 0), the ier axiizatio proble over ν ca choose ν j as if h j (x) > 0 (or ν j as if h j (x) < 0) to cause the ier axiizatio blow off to, therefore beig igored by the outer iiizatio over x For ay x that does ot violate ay of the equality or iequality costraits, i the ier axiizatio proble over ν, the expressio ν jh j (x) evaluates to 0 o atter what the value of ν is, ad i the ier axiizatio proble over λ, the expressio λ i f i (x) ca at axiu be 0, because λ i is costraied to be o-egative, ad f i (x) is o-positive. Therefore, at best, the axiizatio proble sets λ i f i (x) = 0, ad ax L (x,λ,ν) = f 0(x) λ 0,ν I its full for, the objective L (x,λ,ν) is called the Lagragia, ad it takes ito accout the ucostraied set of prial variables x R d, the costraied set of dual variables λ R correspodig to the iequality costraits, ad the ucostraied set of dual variables ν R correspodig to the equality costraits. Note that our dual variables λ i are i fact costraied, so ultiately we were ot able to tur the origial optiizatio proble ito a ucostraied oe, but our costraits are uch sipler tha before. (4) CS 189, Fall 2017, Note 18 3

4 The dual of this optiizatio proble is still over the sae optiizatio objective, except that ow we swap the order of the axiizatio of the dual variables ad the iiizatio of the prial variables. d = ax i L (x,λ,ν) = ax g(λ,ν) λ 0,ν x λ 0,ν The dual is effectively a axiizatio proble (over the dual variables) of g(λ,ν) = i x L (x,λ,ν). 2 Strog Duality ad KKT Coditios It is always true that the solutio to the prial proble is at least as large as the solutio to the dual proble: p d (5) This coditio is kow as weak duality. Proof. We kow that x,λ 0,ν ax L (x, λ, ν) L (x,λ,ν) il (,λ,ν) More copactly, x,λ 0,ν ax L (x, λ, ν) il (,λ,ν) Sice this is true for all x,λ 0,ν this is true i particular whe we set x = argi ax L (, λ, ν) ad λ,ν = argax il (, λ, ν) We therefore kow that p = i ax L (, λ, ν) ax il (, λ, ν) = d The differece p d is kow as the duality gap. I the case of strog duality, the duality gap is 0. That is, we ca swap the order of the iiizatio ad axiizatio ad up with the sae optial value: p = d (6) There are several useful theores detailig the existece of strog duality, such as Slater s theore, which states that if the prial proble is covex, ad there exists a x that ca strictly eet the iequality costraits ad eet the equality costraits, the strog duality holds. Give that strog duality holds, the Karush-Kuh-Tucker (KKT) coditios ca help us fid the solutios to the dual variables of the optiizatio proble. The KKT coditios are coposed of: CS 189, Fall 2017, Note 18 4

5 1. Prial feasibility (iequalities) f i (x) 0, i [1,] 2. Prial feasibility (equalities) 3. Dual feasibility 4. Copleetary Slackess h j (x) = 0, j [1,] λ i 0, i [1,] λ i f i (x) = 0, i [1,] 5. Statioarity x f 0 (x) + λ i x f i (x) + ν j x h j (x) = 0 Let s see how the KKT coditios relate to strog duality. Theore 2.1. If x ad λ,ν are the prial ad dual solutios respectively, with zero duality gap (i.e. strog duality holds), the x,λ,ν also satisfy the KKT coditios. Proof. KKT coditios 1, 2, 3 are trivially true, because the prial solutio x ust satisfy the prial costraits, ad the dual solutio λ,ν ust satisfy the dual costraits. Now, let s prove coditios 4 ad 5. We kow that sice strog duality holds, we ca say that p = f 0 (x ) = g(λ,ν ) = d = il (x,λ,ν ) x L (x,λ,ν ) = f 0 (x ) + = f 0 (x ) + f 0 (x ) λ i f i (x ) + λ i f i (x ) ν j h j (x ) I the fourth step, we ca cacel the ters ivolvig h j (x ) because we kow that the prial solutio ust satisfy h j (x ) = 0. I the fifth step, we kow that λi f i (x ) 0, because λi 0 i order to satisfy the dual costraits, ad f i (x ) 0 i order to satisfy the prial costraits. Sice we established that f 0 (x ) = i x L (x,λ,ν ) L (x,λ,ν ) f 0 (x ), we kow that all of the iequalities hold with iequality ad therefore L (x,λ,ν ) = i x L (x,λ,ν ). This iplies KKT coditio 5 (statioarity), that x f 0 (x ) + λi x f i (x ) + ν j xh j (x ) = 0 CS 189, Fall 2017, Note 18 5

6 Fially, ote that due to the equality f 0 (x )+ λ i f i (x ) = f 0 (x ), we kow that λ i f i (x ) = 0. This cobied with the fact that i λi f i (x ) 0, establishes KKT coditio 4 (copleetary slackess): λi f i (x ) = 0, i [1,] The theore above establishes that i the presece of strog duality, if the solutios are optial, the they satisfy the KKT coditios. A stateet that is alost, but ot quite the coverse, is also true. Theore 2.2. If the prial proble is covex, ad if x ad λ,ν satisfy the KKT coditios, the they are the optial solutios to the prial ad dual probles, respectively. Proof. If x ad λ,ν satisfy KKT coditios 1, 2, 3 we kow that they are at least feasible for the prial ad dual proble. Fro the KKT statioarity coditio we kow that x f 0 (x ) + λi x f i (x ) + ν j xh j (x ) = 0 Sice the prial proble is covex, we kow that L (x,λ,ν) is covex i x, ad if the gradiet of L (x,λ,ν ) at x is 0, we kow that x = i x L (x,λ,ν ) Therefore, we kow that the optial prial values for the prial proble optiize the ier optiizatio proble of the dual proble, ad g(λ,ν ) = f 0 (x ) + λi f i (x ) + ν j h j (x ) By the prial feasibility coditios for h j (x) ad the copleetary slackess coditio, we kow that g(λ,ν ) = f 0 (x ) Now, all we have to do is to prove that x ad λ,ν are prial ad dual optial, respectively. Note that sice weak duality always holds, we kow that p d = ax λ 0,ν g(λ,ν) g( λ, ν), Sice we kow that p g(λ,ν), we ca also say that f 0 (x) p f 0 (x) g(λ,ν) Ad if we have that f 0 (x ) = g(λ,ν ) as we deduced earlier, the λ 0, ν f 0 (x ) p f 0 (x ) g(λ,ν ) = 0 = p = f 0 (x ) = g(λ,ν ) = d Therefore, we have prove that x ad λ,ν are prial ad dual optial, respectively. Eve though we did ot iitially assue that strog duality holds, we evetually arrived at the coclusio that strog duality does ideed hold. CS 189, Fall 2017, Note 18 6

7 Let s pause for a secod to uderstad what we ve foud so far. Give a optiizatio proble, its prial proble is a optiizatio proble over the prial variables, ad its dual proble is a optiizatio proble over the dual variables. If we kow that strog duality holds, the we ca either solve the prial or dual proble ad ed up with the sae optial value. Whe solvig for the dual variables, if we kow that the prial proble is covex i the prial variables, the we use the KKT coditios to fid the optial dual variables. We shall do just that, i our discussio of SVMs. 3 The Dual of SVMs Now, let s apply our kowledge of duality to fid the dual of the soft-argi SVM optiizatio proble. i f (w,t,ξ ) {}}{ 1 2 w 2 +C ξ i i w,t,ξ (1 ξ i ) y i (w T x i t) 0 ad ξ i 0 }{{} g(w,t,ξ ) 0 Let s idetify the prial ad dual variables for the SVM proble. We will have Prial variables w, t, ad ξ i Dual variables α i correspodig to each costrait of the for y i (w T x i t) 1 ξ i Dual variables β i correspodig to each costrait of the for ξ i 0 For the purposes of otatio, ote that we are usig α ad β i place of λ, ad there are dual variables correspodig to ν because there are o equality costraits. The lagragia for the SVM proble is: L (w,t,ξ,α,β) = 1 2 w 2 +C Thus, the dual is: = 1 2 w 2 1 ax g(α,β) = i α,β 0 w,t,ξ 2 w 2 ξ i + α i y i (w T x i t) + α i ((1 ξ i ) y i (w T x i t)) + α i y i (w T x i t) + α i + α i + β i ( ξ i ) (C α i β i )ξ i (7) (C α i β i )ξ i (8) Let s use the KKT coditios to fid the optial dual variables. Verify that the prial proble is covex i the prial variables. We kow that fro the statioarity coditios, evaluated at the optial dual values α ad β, ad the optial prial values w,t,ξ i : L w i = L t = L ξ i = 0 CS 189, Fall 2017, Note 18 7

8 w L = w α i y ix i = 0 = w = α i y ix i. This tells us that w is goig to be a weighted cobiatio of the positive-class x i s ad egative-class x i s. L t = α i y i = 0. This tells us that the weights α i will be equally distributed aog positive- ad egative- class traiig poits. = C αi βi = 0 = 0 αi C. This tells us that the weights αi are restricted to beig less tha the hyperparaeter C. L ξ i Verify that the other KKT also hold. Usig these observatios, we ca eliiate soe ters of the dual proble. L (w,t,ξ,α,β ) = 1 2 w 2 = 1 2 w 2 = 1 2 w 2 α i y i (w T x i t) + α i y i (w T x i ) +t α i y i (w T x i ) + α i + αi y i + }{{} =0 αi (C α i β α i + i )ξ i (C α i β i )ξ i } {{ } =0 We ca also rewrite all the optial prial variables w,t,ξ i ters of the optial dual variables α i : g(α,β ) = i w,t,ξ L (w,t,ξ,α,β ) = L (w,t,ξ,α,β ) = 1 2 α i y i x i 2 = 1 2 α i y i x i 2 = α T α T Qα α i y i (( (α i y i x T i ( where Q i j = y i (x T i x j)y j (ad Q = (diag y)xx T (diag y)). α j y j x j ) T x i ) + α j y j x j )) + αi αi Now, we ca write the fial for of the dual, which is oly i ters of α (ad x ad y): ax α αt αt Qα α i y i = 0 0 α i C CS 189, Fall 2017, Note 18 8

9 ax α s.t. α T αt Qα α i y i = 0 0 α i C i = 1,..., (9) 3.1 Geoetric ituitio We ve see that the optial value of the dual proble i ters of α is equivalet to the optial value of the prial proble i ters of w, t, ad ξ. But what do these dual values α i eve ea? That s a good questio! Recall the followig KKT coditios that are eforced: Statioarity Copleetary slackess C α i β i = 0 α i ((1 ξ i ) y i (w T x i t )) = 0 β i ξ i = 0 Here are soe oteworthy relatioships betwee α i ad the properties of the SVMs: Case 1: αi = 0. I this case, we kow βi = C, which is ozero, ad therefore ξi = 0. That is, if for poit i we have that αi = 0 by the dual proble, the we kow that there is o slack give to this poit. Lookig at the other copleetary slackess coditio, this akes sese because if αi = 0, the y i (w T x i t ) (1 ξi ) ay be ay value, ad if we re iiizig the su of our ξ i s, we should have ξi = 0. Case 2: α i is ozero. If this is the case, the we kow β i = C α i 0 Case 2.1: αi = C. If this is the case, the we kow βi = 0, ad therefore ξi ay be exactly 0 or ozero. Case 2.2: αi lies betwee 0 ad C. I this case, the βi is ozero ad ξi = 0. But this is differet fro Case 1 because with αi ozero, we ca divide by αi i the copleetary slackess coditio ad arrive at the fact that y i (w T x i t) 1 = 0 = y i (w T x i ) = t, which eas x i lies exactly o the argi deteried by w ad t. Lastly, let s recostruct the optial prial values w,t,ξi fro the optial dual values α : w = α i y i x i t = ea(w T x i i : 0 < αi < C) { ξi 1 y i (w T x i t ) if αi = C, = 0 otherwise (10) CS 189, Fall 2017, Note 18 9

Linear Support Vector Machines

Linear Support Vector Machines Liear Support Vector Machies David S. Roseberg The Support Vector Machie For a liear support vector machie (SVM), we use the hypothesis space of affie fuctios F = { f(x) = w T x + b w R d, b R } ad evaluate

More information

Introduction to Optimization, DIKU Monday 19 November David Pisinger. Duality, motivation

Introduction to Optimization, DIKU Monday 19 November David Pisinger. Duality, motivation Itroductio to Optiizatio, DIKU 007-08 Moday 9 Noveber David Pisiger Lecture, Duality ad sesitivity aalysis Duality, shadow prices, sesitivity aalysis, post-optial aalysis, copleetary slackess, KKT optiality

More information

1 Duality revisited. AM 221: Advanced Optimization Spring 2016

1 Duality revisited. AM 221: Advanced Optimization Spring 2016 AM 22: Advaced Optimizatio Sprig 206 Prof. Yaro Siger Sectio 7 Wedesday, Mar. 9th Duality revisited I this sectio, we will give a slightly differet perspective o duality. optimizatio program: f(x) x R

More information

LP in Standard and Slack Forms

LP in Standard and Slack Forms LP i Stadard ad Slack Fors ax j=1 s.t. j=1 c j a ij b i for i=1, 2,..., 0 for j=1, 2,..., z = 0 j=1 c j x i = b i j=1 a ij for i=1, 2,..., Auxiliary Liear Progra L: LP i stadard for: ax j=1 L aux : Auxiliary

More information

Statistics and Data Analysis in MATLAB Kendrick Kay, February 28, Lecture 4: Model fitting

Statistics and Data Analysis in MATLAB Kendrick Kay, February 28, Lecture 4: Model fitting Statistics ad Data Aalysis i MATLAB Kedrick Kay, kedrick.kay@wustl.edu February 28, 2014 Lecture 4: Model fittig 1. The basics - Suppose that we have a set of data ad suppose that we have selected the

More information

Machine Learning Theory Tübingen University, WS 2016/2017 Lecture 12

Machine Learning Theory Tübingen University, WS 2016/2017 Lecture 12 Machie Learig Theory Tübige Uiversity, WS 06/07 Lecture Tolstikhi Ilya Abstract I this lecture we derive risk bouds for kerel methods. We will start by showig that Soft Margi kerel SVM correspods to miimizig

More information

Summer MA Lesson 13 Section 1.6, Section 1.7 (part 1)

Summer MA Lesson 13 Section 1.6, Section 1.7 (part 1) Suer MA 1500 Lesso 1 Sectio 1.6, Sectio 1.7 (part 1) I Solvig Polyoial Equatios Liear equatio ad quadratic equatios of 1 variable are specific types of polyoial equatios. Soe polyoial equatios of a higher

More information

Support vector machine revisited

Support vector machine revisited 6.867 Machie learig, lecture 8 (Jaakkola) 1 Lecture topics: Support vector machie ad kerels Kerel optimizatio, selectio Support vector machie revisited Our task here is to first tur the support vector

More information

Boosting. Professor Ameet Talwalkar. Professor Ameet Talwalkar CS260 Machine Learning Algorithms March 1, / 32

Boosting. Professor Ameet Talwalkar. Professor Ameet Talwalkar CS260 Machine Learning Algorithms March 1, / 32 Boostig Professor Ameet Talwalkar Professor Ameet Talwalkar CS260 Machie Learig Algorithms March 1, 2017 1 / 32 Outlie 1 Admiistratio 2 Review of last lecture 3 Boostig Professor Ameet Talwalkar CS260

More information

Name Period ALGEBRA II Chapter 1B and 2A Notes Solving Inequalities and Absolute Value / Numbers and Functions

Name Period ALGEBRA II Chapter 1B and 2A Notes Solving Inequalities and Absolute Value / Numbers and Functions Nae Period ALGEBRA II Chapter B ad A Notes Solvig Iequalities ad Absolute Value / Nubers ad Fuctios SECTION.6 Itroductio to Solvig Equatios Objectives: Write ad solve a liear equatio i oe variable. Solve

More information

Supplementary Material

Supplementary Material Suppleetary Material Wezhuo Ya a0096049@us.edu.s Departet of Mechaical Eieeri, Natioal Uiversity of Siapore, Siapore 117576 Hua Xu pexuh@us.edu.s Departet of Mechaical Eieeri, Natioal Uiversity of Siapore,

More information

A PROBABILITY PROBLEM

A PROBABILITY PROBLEM A PROBABILITY PROBLEM A big superarket chai has the followig policy: For every Euros you sped per buy, you ear oe poit (suppose, e.g., that = 3; i this case, if you sped 8.45 Euros, you get two poits,

More information

Lecture 10: Bounded Linear Operators and Orthogonality in Hilbert Spaces

Lecture 10: Bounded Linear Operators and Orthogonality in Hilbert Spaces Lecture : Bouded Liear Operators ad Orthogoality i Hilbert Spaces 34 Bouded Liear Operator Let ( X, ), ( Y, ) i i be ored liear vector spaces ad { } X Y The, T is said to be bouded if a real uber c such

More information

Introduction to Machine Learning Spring 2018 Note Duality. 1.1 Primal and Dual Problem

Introduction to Machine Learning Spring 2018 Note Duality. 1.1 Primal and Dual Problem CS 189 Introduction to Machine Learning Spring 2018 Note 22 1 Duality As we have seen in our discussion of kernels, ridge regression can be viewed in two ways: (1) an optimization problem over the weights

More information

Optimal Estimator for a Sample Set with Response Error. Ed Stanek

Optimal Estimator for a Sample Set with Response Error. Ed Stanek Optial Estiator for a Saple Set wit Respose Error Ed Staek Itroductio We develop a optial estiator siilar to te FP estiator wit respose error tat was cosidered i c08ed63doc Te first 6 pages of tis docuet

More information

Keywords: duality, saddle point, complementary slackness, Karush-Kuhn-Tucker conditions, perturbation function, supporting functions.

Keywords: duality, saddle point, complementary slackness, Karush-Kuhn-Tucker conditions, perturbation function, supporting functions. DUALITY THEORY Jørge Tid Uiversity of Copehage, Deark. Keywords: duality, saddle poit, copleetary slackess, KarushKuhTucker coditios, perturbatio fuctio, supportig fuctios. Cotets 1. Itroductio 2. Covex

More information

6.867 Machine learning, lecture 13 (Jaakkola)

6.867 Machine learning, lecture 13 (Jaakkola) Lecture topics: Boostig, argi, ad gradiet descet copleity of classifiers, geeralizatio Boostig Last tie we arrived at a boostig algorith for sequetially creatig a eseble of base classifiers. Our base classifiers

More information

11 KERNEL METHODS From Feature Combinations to Kernels. φ(x) = 1, 2x 1, 2x 2, 2x 3,..., 2x D, Learning Objectives:

11 KERNEL METHODS From Feature Combinations to Kernels. φ(x) = 1, 2x 1, 2x 2, 2x 3,..., 2x D, Learning Objectives: 11 KERNEL METHODS May who have had a opportuity of kowig ay ore about atheatics cofuse it with arithetic, ad cosider it a arid sciece. I reality, however, it is a sciece which requires a great aout of

More information

CSCI567 Machine Learning (Fall 2014)

CSCI567 Machine Learning (Fall 2014) CSCI567 Machie Learig (Fall 2014) Drs. Sha & Liu {feisha,yaliu.cs}@usc.edu October 9, 2014 Drs. Sha & Liu ({feisha,yaliu.cs}@usc.edu) CSCI567 Machie Learig (Fall 2014) October 9, 2014 1 / 49 Outlie Admiistratio

More information

The Binomial Multi-Section Transformer

The Binomial Multi-Section Transformer 4/15/2010 The Bioial Multisectio Matchig Trasforer preset.doc 1/24 The Bioial Multi-Sectio Trasforer Recall that a ulti-sectio atchig etwork ca be described usig the theory of sall reflectios as: where:

More information

ECE 901 Lecture 4: Estimation of Lipschitz smooth functions

ECE 901 Lecture 4: Estimation of Lipschitz smooth functions ECE 9 Lecture 4: Estiatio of Lipschitz sooth fuctios R. Nowak 5/7/29 Cosider the followig settig. Let Y f (X) + W, where X is a rado variable (r.v.) o X [, ], W is a r.v. o Y R, idepedet of X ad satisfyig

More information

Seunghee Ye Ma 8: Week 5 Oct 28

Seunghee Ye Ma 8: Week 5 Oct 28 Week 5 Summary I Sectio, we go over the Mea Value Theorem ad its applicatios. I Sectio 2, we will recap what we have covered so far this term. Topics Page Mea Value Theorem. Applicatios of the Mea Value

More information

5.6 Binomial Multi-section Matching Transformer

5.6 Binomial Multi-section Matching Transformer 4/14/21 5_6 Bioial Multisectio Matchig Trasforers 1/1 5.6 Bioial Multi-sectio Matchig Trasforer Readig Assiget: pp. 246-25 Oe way to axiize badwidth is to costruct a ultisectio Γ f that is axially flat.

More information

Support Vector Machines and Kernel Methods

Support Vector Machines and Kernel Methods Support Vector Machies ad Kerel Methods Daiel Khashabi Fall 202 Last Update: September 26, 206 Itroductio I Support Vector Machies the goal is to fid a separator betwee data which has the largest margi,

More information

X. Perturbation Theory

X. Perturbation Theory X. Perturbatio Theory I perturbatio theory, oe deals with a ailtoia that is coposed Ĥ that is typically exactly solvable of two pieces: a referece part ad a perturbatio ( Ĥ ) that is assued to be sall.

More information

6.3 Testing Series With Positive Terms

6.3 Testing Series With Positive Terms 6.3. TESTING SERIES WITH POSITIVE TERMS 307 6.3 Testig Series With Positive Terms 6.3. Review of what is kow up to ow I theory, testig a series a i for covergece amouts to fidig the i= sequece of partial

More information

and then substitute this into the second equation to get 5(11 4 y) 3y

and then substitute this into the second equation to get 5(11 4 y) 3y Math E-b Lecture # Notes The priary focus of this week s lecture is a systeatic way of solvig ad uderstadig systes of liear equatios algebraically, geoetrically, ad logically. Eaple #: Solve the syste

More information

Summary: Congruences. j=1. 1 Here we use the Mathematica syntax for the function. In Maple worksheets, the function

Summary: Congruences. j=1. 1 Here we use the Mathematica syntax for the function. In Maple worksheets, the function Summary: Cogrueces j whe divided by, ad determiig the additive order of a iteger mod. As described i the Prelab sectio, cogrueces ca be thought of i terms of coutig with rows, ad for some questios this

More information

The Hypergeometric Coupon Collection Problem and its Dual

The Hypergeometric Coupon Collection Problem and its Dual Joural of Idustrial ad Systes Egieerig Vol., o., pp -7 Sprig 7 The Hypergeoetric Coupo Collectio Proble ad its Dual Sheldo M. Ross Epstei Departet of Idustrial ad Systes Egieerig, Uiversity of Souther

More information

f(1), and so, if f is continuous, f(x) = f(1)x.

f(1), and so, if f is continuous, f(x) = f(1)x. 2.2.35: Let f be a additive fuctio. i Clearly fx = fx ad therefore f x = fx for all Z+ ad x R. Hece, for ay, Z +, f = f, ad so, if f is cotiuous, fx = fx. ii Suppose that f is bouded o soe o-epty ope set.

More information

Jacobi symbols. p 1. Note: The Jacobi symbol does not necessarily distinguish between quadratic residues and nonresidues. That is, we could have ( a

Jacobi symbols. p 1. Note: The Jacobi symbol does not necessarily distinguish between quadratic residues and nonresidues. That is, we could have ( a Jacobi sybols efiitio Let be a odd positive iteger If 1, the Jacobi sybol : Z C is the costat fuctio 1 1 If > 1, it has a decopositio ( as ) a product of (ot ecessarily distict) pries p 1 p r The Jacobi

More information

The Growth of Functions. Theoretical Supplement

The Growth of Functions. Theoretical Supplement The Growth of Fuctios Theoretical Supplemet The Triagle Iequality The triagle iequality is a algebraic tool that is ofte useful i maipulatig absolute values of fuctios. The triagle iequality says that

More information

Physics 116A Solutions to Homework Set #1 Winter Boas, problem Use equation 1.8 to find a fraction describing

Physics 116A Solutions to Homework Set #1 Winter Boas, problem Use equation 1.8 to find a fraction describing Physics 6A Solutios to Homework Set # Witer 0. Boas, problem. 8 Use equatio.8 to fid a fractio describig 0.694444444... Start with the formula S = a, ad otice that we ca remove ay umber of r fiite decimals

More information

Answer Key, Problem Set 1, Written

Answer Key, Problem Set 1, Written Cheistry 1 Mies, Sprig, 018 Aswer Key, Proble Set 1, Writte 1. 14.3;. 14.34 (add part (e): Estiate / calculate the iitial rate of the reactio); 3. NT1; 4. NT; 5. 14.37; 6. 14.39; 7. 14.41; 8. NT3; 9. 14.46;

More information

5.6 Binomial Multi-section Matching Transformer

5.6 Binomial Multi-section Matching Transformer 4/14/2010 5_6 Bioial Multisectio Matchig Trasforers 1/1 5.6 Bioial Multi-sectio Matchig Trasforer Readig Assiget: pp. 246-250 Oe way to axiize badwidth is to costruct a ultisectio Γ f that is axially flat.

More information

18.657: Mathematics of Machine Learning

18.657: Mathematics of Machine Learning 8.657: Mathematics of Machie Learig Lecturer: Philippe Rigollet Lecture 0 Scribe: Ade Forrow Oct. 3, 05 Recall the followig defiitios from last time: Defiitio: A fuctio K : X X R is called a positive symmetric

More information

COMP 2804 Solutions Assignment 1

COMP 2804 Solutions Assignment 1 COMP 2804 Solutios Assiget 1 Questio 1: O the first page of your assiget, write your ae ad studet uber Solutio: Nae: Jaes Bod Studet uber: 007 Questio 2: I Tic-Tac-Toe, we are give a 3 3 grid, cosistig

More information

Lecture 19. Curve fitting I. 1 Introduction. 2 Fitting a constant to measured data

Lecture 19. Curve fitting I. 1 Introduction. 2 Fitting a constant to measured data Lecture 9 Curve fittig I Itroductio Suppose we are preseted with eight poits of easured data (x i, y j ). As show i Fig. o the left, we could represet the uderlyig fuctio of which these data are saples

More information

Introduction to Machine Learning DIS10

Introduction to Machine Learning DIS10 CS 189 Fall 017 Itroductio to Machie Learig DIS10 1 Fu with Lagrage Multipliers (a) Miimize the fuctio such that f (x,y) = x + y x + y = 3. Solutio: The Lagragia is: L(x,y,λ) = x + y + λ(x + y 3) Takig

More information

42 Dependence and Bases

42 Dependence and Bases 42 Depedece ad Bases The spa s(a) of a subset A i vector space V is a subspace of V. This spa ay be the whole vector space V (we say the A spas V). I this paragraph we study subsets A of V which spa V

More information

6.867 Machine learning, lecture 7 (Jaakkola) 1

6.867 Machine learning, lecture 7 (Jaakkola) 1 6.867 Machie learig, lecture 7 (Jaakkola) 1 Lecture topics: Kerel form of liear regressio Kerels, examples, costructio, properties Liear regressio ad kerels Cosider a slightly simpler model where we omit

More information

Polynomials with Rational Roots that Differ by a Non-zero Constant. Generalities

Polynomials with Rational Roots that Differ by a Non-zero Constant. Generalities Polyomials with Ratioal Roots that Differ by a No-zero Costat Philip Gibbs The problem of fidig two polyomials P(x) ad Q(x) of a give degree i a sigle variable x that have all ratioal roots ad differ by

More information

Machine Learning Brett Bernstein

Machine Learning Brett Bernstein Machie Learig Brett Berstei Week 2 Lecture: Cocept Check Exercises Starred problems are optioal. Excess Risk Decompositio 1. Let X = Y = {1, 2,..., 10}, A = {1,..., 10, 11} ad suppose the data distributio

More information

RADICAL EXPRESSION. If a and x are real numbers and n is a positive integer, then x is an. n th root theorems: Example 1 Simplify

RADICAL EXPRESSION. If a and x are real numbers and n is a positive integer, then x is an. n th root theorems: Example 1 Simplify Example 1 Simplify 1.2A Radical Operatios a) 4 2 b) 16 1 2 c) 16 d) 2 e) 8 1 f) 8 What is the relatioship betwee a, b, c? What is the relatioship betwee d, e, f? If x = a, the x = = th root theorems: RADICAL

More information

1. By using truth tables prove that, for all statements P and Q, the statement

1. By using truth tables prove that, for all statements P and Q, the statement Author: Satiago Salazar Problems I: Mathematical Statemets ad Proofs. By usig truth tables prove that, for all statemets P ad Q, the statemet P Q ad its cotrapositive ot Q (ot P) are equivalet. I example.2.3

More information

Homework Set #3 - Solutions

Homework Set #3 - Solutions EE 15 - Applicatios of Covex Optimizatio i Sigal Processig ad Commuicatios Dr. Adre Tkaceko JPL Third Term 11-1 Homework Set #3 - Solutios 1. a) Note that x is closer to x tha to x l i the Euclidea orm

More information

( 1) n (4x + 1) n. n=0

( 1) n (4x + 1) n. n=0 Problem 1 (10.6, #). Fid the radius of covergece for the series: ( 1) (4x + 1). For what values of x does the series coverge absolutely, ad for what values of x does the series coverge coditioally? Solutio.

More information

Automated Proofs for Some Stirling Number Identities

Automated Proofs for Some Stirling Number Identities Autoated Proofs for Soe Stirlig Nuber Idetities Mauel Kauers ad Carste Scheider Research Istitute for Sybolic Coputatio Johaes Kepler Uiversity Altebergerstraße 69 A4040 Liz, Austria Subitted: Sep 1, 2007;

More information

Machine Learning Theory Tübingen University, WS 2016/2017 Lecture 11

Machine Learning Theory Tübingen University, WS 2016/2017 Lecture 11 Machie Learig Theory Tübige Uiversity, WS 06/07 Lecture Tolstikhi Ilya Abstract We will itroduce the otio of reproducig kerels ad associated Reproducig Kerel Hilbert Spaces (RKHS). We will cosider couple

More information

10-701/ Machine Learning Mid-term Exam Solution

10-701/ Machine Learning Mid-term Exam Solution 0-70/5-78 Machie Learig Mid-term Exam Solutio Your Name: Your Adrew ID: True or False (Give oe setece explaatio) (20%). (F) For a cotiuous radom variable x ad its probability distributio fuctio p(x), it

More information

Machine Learning for Data Science (CS 4786)

Machine Learning for Data Science (CS 4786) Machie Learig for Data Sciece CS 4786) Lecture & 3: Pricipal Compoet Aalysis The text i black outlies high level ideas. The text i blue provides simple mathematical details to derive or get to the algorithm

More information

Orthogonal Functions

Orthogonal Functions Royal Holloway Uiversity of odo Departet of Physics Orthogoal Fuctios Motivatio Aalogy with vectors You are probably failiar with the cocept of orthogoality fro vectors; two vectors are orthogoal whe they

More information

The Binomial Multi- Section Transformer

The Binomial Multi- Section Transformer 4/4/26 The Bioial Multisectio Matchig Trasforer /2 The Bioial Multi- Sectio Trasforer Recall that a ulti-sectio atchig etwork ca be described usig the theory of sall reflectios as: where: ( ω ) = + e +

More information

CSCI567 Machine Learning (Fall 2014)

CSCI567 Machine Learning (Fall 2014) CSCI567 Machie Learig (Fall 2014) Drs. Sha & Liu {feisha,yaliu.cs}@usc.edu October 14, 2014 Drs. Sha & Liu ({feisha,yaliu.cs}@usc.edu) CSCI567 Machie Learig (Fall 2014) October 14, 2014 1 / 49 Outlie Admiistratio

More information

19.1 The dictionary problem

19.1 The dictionary problem CS125 Lecture 19 Fall 2016 19.1 The dictioary proble Cosider the followig data structural proble, usually called the dictioary proble. We have a set of ites. Each ite is a (key, value pair. Keys are i

More information

Bertrand s postulate Chapter 2

Bertrand s postulate Chapter 2 Bertrad s postulate Chapter We have see that the sequece of prie ubers, 3, 5, 7,... is ifiite. To see that the size of its gaps is ot bouded, let N := 3 5 p deote the product of all prie ubers that are

More information

Math 113, Calculus II Winter 2007 Final Exam Solutions

Math 113, Calculus II Winter 2007 Final Exam Solutions Math, Calculus II Witer 7 Fial Exam Solutios (5 poits) Use the limit defiitio of the defiite itegral ad the sum formulas to compute x x + dx The check your aswer usig the Evaluatio Theorem Solutio: I this

More information

Convergence of random variables. (telegram style notes) P.J.C. Spreij

Convergence of random variables. (telegram style notes) P.J.C. Spreij Covergece of radom variables (telegram style otes).j.c. Spreij this versio: September 6, 2005 Itroductio As we kow, radom variables are by defiitio measurable fuctios o some uderlyig measurable space

More information

Chapter 4. Fourier Series

Chapter 4. Fourier Series Chapter 4. Fourier Series At this poit we are ready to ow cosider the caoical equatios. Cosider, for eample the heat equatio u t = u, < (4.) subject to u(, ) = si, u(, t) = u(, t) =. (4.) Here,

More information

A string of not-so-obvious statements about correlation in the data. (This refers to the mechanical calculation of correlation in the data.

A string of not-so-obvious statements about correlation in the data. (This refers to the mechanical calculation of correlation in the data. STAT-UB.003 NOTES for Wedesday 0.MAY.0 We will use the file JulieApartet.tw. We ll give the regressio of Price o SqFt, show residual versus fitted plot, save residuals ad fitted. Give plot of (Resid, Price,

More information

ECE Spring Prof. David R. Jackson ECE Dept. Notes 20

ECE Spring Prof. David R. Jackson ECE Dept. Notes 20 ECE 6341 Sprig 016 Prof. David R. Jackso ECE Dept. Notes 0 1 Spherical Wave Fuctios Cosider solvig ψ + k ψ = 0 i spherical coordiates z φ θ r y x Spherical Wave Fuctios (cot.) I spherical coordiates we

More information

Machine Learning for Data Science (CS 4786)

Machine Learning for Data Science (CS 4786) Machie Learig for Data Sciece CS 4786) Lecture 9: Pricipal Compoet Aalysis The text i black outlies mai ideas to retai from the lecture. The text i blue give a deeper uderstadig of how we derive or get

More information

Ma 530 Introduction to Power Series

Ma 530 Introduction to Power Series Ma 530 Itroductio to Power Series Please ote that there is material o power series at Visual Calculus. Some of this material was used as part of the presetatio of the topics that follow. What is a Power

More information

w (1) ˆx w (1) x (1) /ρ and w (2) ˆx w (2) x (2) /ρ.

w (1) ˆx w (1) x (1) /ρ and w (2) ˆx w (2) x (2) /ρ. 2 5. Weighted umber of late jobs 5.1. Release dates ad due dates: maximimizig the weight of o-time jobs Oce we add release dates, miimizig the umber of late jobs becomes a sigificatly harder problem. For

More information

Recurrence Relations

Recurrence Relations Recurrece Relatios Aalysis of recursive algorithms, such as: it factorial (it ) { if (==0) retur ; else retur ( * factorial(-)); } Let t be the umber of multiplicatios eeded to calculate factorial(). The

More information

Fourier Series and the Wave Equation

Fourier Series and the Wave Equation Fourier Series ad the Wave Equatio We start with the oe-dimesioal wave equatio u u =, x u(, t) = u(, t) =, ux (,) = f( x), u ( x,) = This represets a vibratig strig, where u is the displacemet of the strig

More information

11.6 Absolute Convergence and the Ratio and Root Tests

11.6 Absolute Convergence and the Ratio and Root Tests .6 Absolute Covergece ad the Ratio ad Root Tests The most commo way to test for covergece is to igore ay positive or egative sigs i a series, ad simply test the correspodig series of positive terms. Does

More information

Some examples of vector spaces

Some examples of vector spaces Roberto s Notes o Liear Algebra Chapter 11: Vector spaces Sectio 2 Some examples of vector spaces What you eed to kow already: The te axioms eeded to idetify a vector space. What you ca lear here: Some

More information

Math 4707 Spring 2018 (Darij Grinberg): midterm 2 page 1. Math 4707 Spring 2018 (Darij Grinberg): midterm 2 with solutions [preliminary version]

Math 4707 Spring 2018 (Darij Grinberg): midterm 2 page 1. Math 4707 Spring 2018 (Darij Grinberg): midterm 2 with solutions [preliminary version] Math 4707 Sprig 08 Darij Griberg: idter page Math 4707 Sprig 08 Darij Griberg: idter with solutios [preliiary versio] Cotets 0.. Coutig first-eve tuples......................... 3 0.. Coutig legal paths

More information

10. Comparative Tests among Spatial Regression Models. Here we revisit the example in Section 8.1 of estimating the mean of a normal random

10. Comparative Tests among Spatial Regression Models. Here we revisit the example in Section 8.1 of estimating the mean of a normal random Part III. Areal Data Aalysis 0. Comparative Tests amog Spatial Regressio Models While the otio of relative likelihood values for differet models is somewhat difficult to iterpret directly (as metioed above),

More information

Section 1 of Unit 03 (Pure Mathematics 3) Algebra

Section 1 of Unit 03 (Pure Mathematics 3) Algebra Sectio 1 of Uit 0 (Pure Mathematics ) Algebra Recommeded Prior Kowledge Studets should have studied the algebraic techiques i Pure Mathematics 1. Cotet This Sectio should be studied early i the course

More information

An alternating series is a series where the signs alternate. Generally (but not always) there is a factor of the form ( 1) n + 1

An alternating series is a series where the signs alternate. Generally (but not always) there is a factor of the form ( 1) n + 1 Calculus II - Problem Solvig Drill 20: Alteratig Series, Ratio ad Root Tests Questio No. of 0 Istructios: () Read the problem ad aswer choices carefully (2) Work the problems o paper as eeded (3) Pick

More information

The Random Walk For Dummies

The Random Walk For Dummies The Radom Walk For Dummies Richard A Mote Abstract We look at the priciples goverig the oe-dimesioal discrete radom walk First we review five basic cocepts of probability theory The we cosider the Beroulli

More information

How to Maximize a Function without Really Trying

How to Maximize a Function without Really Trying How to Maximize a Fuctio without Really Tryig MARK FLANAGAN School of Electrical, Electroic ad Commuicatios Egieerig Uiversity College Dubli We will prove a famous elemetary iequality called The Rearragemet

More information

Math 2784 (or 2794W) University of Connecticut

Math 2784 (or 2794W) University of Connecticut ORDERS OF GROWTH PAT SMITH Math 2784 (or 2794W) Uiversity of Coecticut Date: Mar. 2, 22. ORDERS OF GROWTH. Itroductio Gaiig a ituitive feel for the relative growth of fuctios is importat if you really

More information

Linear Classifiers III

Linear Classifiers III Uiversität Potsdam Istitut für Iformatik Lehrstuhl Maschielles Lere Liear Classifiers III Blaie Nelso, Tobias Scheffer Cotets Classificatio Problem Bayesia Classifier Decisio Liear Classifiers, MAP Models

More information

Pointwise observation of the state given by parabolic system with boundary condition involving multiple time delays

Pointwise observation of the state given by parabolic system with boundary condition involving multiple time delays 1.1515/acsc-216-11 Archives of Cotrol Scieces Volue 26(LXII), 216 No. 2, pages 189 197 Poitwise observatio of the state give by parabolic syste with boudary coditio ivolvig ultiple tie delays ADAM KOWALEWSKI

More information

NICK DUFRESNE. 1 1 p(x). To determine some formulas for the generating function of the Schröder numbers, r(x) = a(x) =

NICK DUFRESNE. 1 1 p(x). To determine some formulas for the generating function of the Schröder numbers, r(x) = a(x) = AN INTRODUCTION TO SCHRÖDER AND UNKNOWN NUMBERS NICK DUFRESNE Abstract. I this article we will itroduce two types of lattice paths, Schröder paths ad Ukow paths. We will examie differet properties of each,

More information

Inverse Matrix. A meaning that matrix B is an inverse of matrix A.

Inverse Matrix. A meaning that matrix B is an inverse of matrix A. Iverse Matrix Two square matrices A ad B of dimesios are called iverses to oe aother if the followig holds, AB BA I (11) The otio is dual but we ofte write 1 B A meaig that matrix B is a iverse of matrix

More information

Sequences I. Chapter Introduction

Sequences I. Chapter Introduction Chapter 2 Sequeces I 2. Itroductio A sequece is a list of umbers i a defiite order so that we kow which umber is i the first place, which umber is i the secod place ad, for ay atural umber, we kow which

More information

Chapter 7. Support Vector Machine

Chapter 7. Support Vector Machine Chapter 7 Support Vector Machie able of Cotet Margi ad support vectors SVM formulatio Slack variables ad hige loss SVM for multiple class SVM ith Kerels Relevace Vector Machie Support Vector Machie (SVM)

More information

) is a square matrix with the property that for any m n matrix A, the product AI equals A. The identity matrix has a ii

) is a square matrix with the property that for any m n matrix A, the product AI equals A. The identity matrix has a ii square atrix is oe that has the sae uber of rows as colus; that is, a atrix. he idetity atrix (deoted by I, I, or [] I ) is a square atrix with the property that for ay atrix, the product I equals. he

More information

Complete Solutions to Supplementary Exercises on Infinite Series

Complete Solutions to Supplementary Exercises on Infinite Series Coplete Solutios to Suppleetary Eercises o Ifiite Series. (a) We eed to fid the su ito partial fractios gives By the cover up rule we have Therefore Let S S A / ad A B B. Covertig the suad / the by usig

More information

SVM for Statisticians

SVM for Statisticians SVM for Statisticias Youyi Fog Fred Hutchiso Cacer Research Istitute November 13, 2011 1 / 21 Primal Problem ad Pealized Loss Fuctio Miimize J over b, β ad ξ uder some costraits J = 1 2 β 2 + C ξ i (1)

More information

Contents Two Sample t Tests Two Sample t Tests

Contents Two Sample t Tests Two Sample t Tests Cotets 3.5.3 Two Saple t Tests................................... 3.5.3 Two Saple t Tests Setup: Two Saples We ow focus o a sceario where we have two idepedet saples fro possibly differet populatios. Our

More information

Summary: CORRELATION & LINEAR REGRESSION. GC. Students are advised to refer to lecture notes for the GC operations to obtain scatter diagram.

Summary: CORRELATION & LINEAR REGRESSION. GC. Students are advised to refer to lecture notes for the GC operations to obtain scatter diagram. Key Cocepts: 1) Sketchig of scatter diagram The scatter diagram of bivariate (i.e. cotaiig two variables) data ca be easily obtaied usig GC. Studets are advised to refer to lecture otes for the GC operatios

More information

Supplemental Material: Proofs

Supplemental Material: Proofs Proof to Theorem Supplemetal Material: Proofs Proof. Let be the miimal umber of traiig items to esure a uique solutio θ. First cosider the case. It happes if ad oly if θ ad Rak(A) d, which is a special

More information

PROPERTIES OF AN EULER SQUARE

PROPERTIES OF AN EULER SQUARE PROPERTIES OF N EULER SQURE bout 0 the mathematicia Leoard Euler discussed the properties a x array of letters or itegers ow kow as a Euler or Graeco-Lati Square Such squares have the property that every

More information

Definitions and Theorems. where x are the decision variables. c, b, and a are constant coefficients.

Definitions and Theorems. where x are the decision variables. c, b, and a are constant coefficients. Defiitios ad Theorems Remember the scalar form of the liear programmig problem, Miimize, Subject to, f(x) = c i x i a 1i x i = b 1 a mi x i = b m x i 0 i = 1,2,, where x are the decisio variables. c, b,

More information

Lecture Outline. 2 Separating Hyperplanes. 3 Banach Mazur Distance An Algorithmist s Toolkit October 22, 2009

Lecture Outline. 2 Separating Hyperplanes. 3 Banach Mazur Distance An Algorithmist s Toolkit October 22, 2009 18.409 A Algorithist s Toolkit October, 009 Lecture 1 Lecturer: Joatha Keler Scribes: Alex Levi (009) 1 Outlie Today we ll go over soe of the details fro last class ad ake precise ay details that were

More information

Geometry Unit 3 Notes Parallel and Perpendicular Lines

Geometry Unit 3 Notes Parallel and Perpendicular Lines Review Cocepts: Equatios of Lies Geoetry Uit Notes Parallel ad Perpedicular Lies Syllabus Objective:. - The studet will differetiate aog parallel, perpedicular, ad skew lies. Lies that DO NOT itersect:

More information

September 2012 C1 Note. C1 Notes (Edexcel) Copyright - For AS, A2 notes and IGCSE / GCSE worksheets 1

September 2012 C1 Note. C1 Notes (Edexcel) Copyright   - For AS, A2 notes and IGCSE / GCSE worksheets 1 September 0 s (Edecel) Copyright www.pgmaths.co.uk - For AS, A otes ad IGCSE / GCSE worksheets September 0 Copyright www.pgmaths.co.uk - For AS, A otes ad IGCSE / GCSE worksheets September 0 Copyright

More information

Notice that this test does not say anything about divergence of an alternating series.

Notice that this test does not say anything about divergence of an alternating series. MATH 572H Sprig 20 Worksheet 7 Topics: absolute ad coditioal covergece; power series. Defiitio. A series b is called absolutely coverget if the series b is coverget. If the series b coverges, while b diverges,

More information

Lecture 6: Integration and the Mean Value Theorem. slope =

Lecture 6: Integration and the Mean Value Theorem. slope = Math 8 Istructor: Padraic Bartlett Lecture 6: Itegratio ad the Mea Value Theorem Week 6 Caltech 202 The Mea Value Theorem The Mea Value Theorem abbreviated MVT is the followig result: Theorem. Suppose

More information

Primes of the form n 2 + 1

Primes of the form n 2 + 1 Itroductio Ladau s Probles are four robles i Nuber Theory cocerig rie ubers: Goldbach s Cojecture: This cojecture states that every ositive eve iteger greater tha ca be exressed as the su of two (ot ecessarily

More information

University of Manitoba, Mathletics 2009

University of Manitoba, Mathletics 2009 Uiversity of Maitoba, Mathletics 009 Sessio 5: Iequalities Facts ad defiitios AM-GM iequality: For a, a,, a 0, a + a + + a (a a a ) /, with equality iff all a i s are equal Cauchy s iequality: For reals

More information

Chapter 7: The z-transform. Chih-Wei Liu

Chapter 7: The z-transform. Chih-Wei Liu Chapter 7: The -Trasform Chih-Wei Liu Outlie Itroductio The -Trasform Properties of the Regio of Covergece Properties of the -Trasform Iversio of the -Trasform The Trasfer Fuctio Causality ad Stability

More information

TR/46 OCTOBER THE ZEROS OF PARTIAL SUMS OF A MACLAURIN EXPANSION A. TALBOT

TR/46 OCTOBER THE ZEROS OF PARTIAL SUMS OF A MACLAURIN EXPANSION A. TALBOT TR/46 OCTOBER 974 THE ZEROS OF PARTIAL SUMS OF A MACLAURIN EXPANSION by A. TALBOT .. Itroductio. A problem i approximatio theory o which I have recetly worked [] required for its solutio a proof that the

More information

SOLUTIONS TO EXAM 3. Solution: Note that this defines two convergent geometric series with respective radii r 1 = 2/5 < 1 and r 2 = 1/5 < 1.

SOLUTIONS TO EXAM 3. Solution: Note that this defines two convergent geometric series with respective radii r 1 = 2/5 < 1 and r 2 = 1/5 < 1. SOLUTIONS TO EXAM 3 Problem Fid the sum of the followig series 2 + ( ) 5 5 2 5 3 25 2 2 This series diverges Solutio: Note that this defies two coverget geometric series with respective radii r 2/5 < ad

More information

MA131 - Analysis 1. Workbook 2 Sequences I

MA131 - Analysis 1. Workbook 2 Sequences I MA3 - Aalysis Workbook 2 Sequeces I Autum 203 Cotets 2 Sequeces I 2. Itroductio.............................. 2.2 Icreasig ad Decreasig Sequeces................ 2 2.3 Bouded Sequeces..........................

More information