Facility location game

Size: px
Start display at page:

Download "Facility location game"

Transcription

1 Mechaism Desig Facility locatio game Recall facility locatio game What are we desigig exactly? Pigzhog Tag Mechaism Desig, Slide

2 Mechaism Desig Facility locatio game We kow the set of types of each player (private locatio) We kow the set of all possible outcomes (locatios of the library) We kow the utility fuctio of each player We eed to desig the set of actios for each player We eed to desig a mappig from a actio profile to a outcome Pigzhog Tag Mechaism Desig, Slide 2

3 Mechaism Desig Mechaism desig I short To complete some basic compoets of a Bayesia game: private ifo give utility fuctio give rules of the game to be desiged Referece book 3, Chapter 0 Eric Maski, Mechaism desig, Nobel Prize 2007 Pigzhog Tag Mechaism Desig, Slide 3

4 Mechaism Desig Eviromet give Exted the ative social choice settig to oe where agets ca t be relied upo to disclose their prefereces hoestly. Start with a Bayesia game, but o actios. Defiitio (Eviromet) A eviromet is a tuple (N,O,,p,u), where N is a fiite set of agets; O is a set of outcomes; = is a set of possible joit type vectors; p is a (commo prior) probability distributio o ; ad u =(u,...,u ),whereu i : O 7! R is the utility fuctio for each player i. Pigzhog Tag Mechaism Desig, Slide 4

5 Mechaism Desig Mechaism Desig Defiitio (Mechaism) A mechaism (for a eviromet (N,O,, p,u)) is a pair(a, g), where A = A A,whereA i is the set of actios available to aget i 2 N; ad g : A 7! (O) maps each actio profile to a distributio over outcomes. (Determiistic mechaisms: A 7! O) Thus, the desiger gets to specify the actio sets for the agets the mappig to outcomes, over which agets have utility ca t chage outcomes; agets prefereces or type spaces Pigzhog Tag Mechaism Desig, Slide 5

6 Mechaism Desig Desig objective The problem is to pick a mechaism that will cause ratioal agets to behave i a desired way, specificallymaximizigthe mechaism desiger s ow objective fuctio, give that each aget holds private iformatio, ukow to the desiger agets play BNE desiger hopes that i BNE, his/her objective is reached Various equivalet ways of lookig at this settig choose the Bayesia game out of a set of possible Bayesia games that maximizes some performace measure desig a game that implemets a particular social choice fuctio C( ) i equilibrium, give that the desiger o loger kows agets prefereces ad the agets might lie Pigzhog Tag Mechaism Desig, Slide 6

7 Mechaism Desig Implemetatio i Domiat Strategies Defiitio (Implemetatio i domiat strategies) Give a eviromet (N,O,, p,u), a mechaism(a, g) is a implemetatio i domiat strategies of a social choice fuctio C :! O if the iduced game has a domiat strategy equilibrium s ( ), ad for ay type profile, wehave g(s ( )) = C( ). E.g., secod-price auctio implemets social-welfare maximizatio i domiat strategies. Pigzhog Tag Mechaism Desig, Slide 7

8 Mechaism Desig Implemetatio i Bayes-Nash equilibrium Defiitio (Bayes Nash implemetatio) Give a Bayesia game settig (N,O,, p,u), a mechaism(a, g) is a implemetatio i Bayes Nash equilibrium of a social choice fuctio C :! O if there exists a Bayes Nash equilibrium s ( ) of the iduced game (N, A,,p,u) such that for ay type profile 2, wehavethatg(s ( )) = C( ). E.g., For symmetric type distributios, first-price auctio implemets social-welfare maximizatio i BNE. Pigzhog Tag Mechaism Desig, Slide 8

9 Mechaism Desig Bayes-Nash Implemetatio Commets Bayes-Nash Equilibrium Problems: there could be more tha oe equilibrium which oe should I expect agets to play? agets could miscoordiate ad play oe of the equilibria Possible Fix: Symmetric Bayes-Nash implemetatio Thm: symmetric Bayesia game has symmetric Bayes-Nash Agai, first-price auctio Pigzhog Tag Mechaism Desig, Slide 9

10 Mechaism Desig Implemetatio Commets We ca require that the desired outcome arises i the oly equilibrium i every equilibrium (strog implemetatio) i at least oe equilibrium (weak implemetatio) Forms of implemetatio: Direct Implemetatio: agets each simultaeously sed a sigle message to the ceter Idirect Implemetatio: agets may sed a sequece of messages; i betwee, iformatio may be (partially) revealed about the messages that were set previously like extesive form Pigzhog Tag Mechaism Desig, Slide 0

11 Revelatio Priciple Impossibility Revelatio Priciple It turs out that ay social choice fuctio that ca be implemeted by some mechaism ca be implemeted by a truthful, direct mechaism! direct mechaism: A i = i truthful, direct mechaism: s i ( i)= i Hold for both DSE ad BNE Cosider a arbitrary (e.g., may be idirect) Pigzhog Tag Mechaism Desig With Urestricted Prefereces, Slide

12 Revelatio Priciple Impossibility Revelatio Priciple type θ type θ strategy s (θ ) M strategy s (θ ) Origial Mechaism outcome It turs out that ay social choice fuctio that ca be implemeted by some mechaism ca be implemeted by a truthful, direct mechaism! direct mechaism: A i = i truthful, direct mechaism: s i ( i)= i Hold for both DSE ad BNE Cosider a arbitrary (e.g., may be idirect) Pigzhog Tag Mechaism Desig With Urestricted Prefereces, Slide

13 Revelatio Priciple Impossibility Revelatio Priciple type θ type θ strategy s (θ ) M strategy s (θ ) Origial Mechaism outcome It turs out that ay social choice fuctio that ca be implemeted by some mechaism ca be implemeted by a truthful, direct mechaism! direct mechaism: A i = i truthful, direct mechaism: s i ( i)= i Hold for both DSE ad BNE Cosider a arbitrary (e.g., may be idirect) Recall that a mechaism defies a game, ad cosider a equilibrium s =(s,...,s ) Pigzhog Tag Mechaism Desig With Urestricted Prefereces, Slide

14 Revelatio Priciple Impossibility Revelatio Priciple type θ type θ strategy s (θ ) M strategy s (θ ) s (s (θ )) M s (s (θ )) Origial Mechaism ( outcome New Mechaism We ca costruct a ew direct mechaism, as show above This mechaism is truthful by exactly the same argumet that s was a equilibrium i the origial mechaism The agets do t have to lie, because the mechaism already lies for them. Pigzhog Tag Mechaism Desig With Urestricted Prefereces, Slide 2

15 Revelatio Priciple Impossibility Discussio of the Revelatio Priciple The set of equilibria is ot always the same i the origial mechaism ad revelatio mechaism of course, we ve show that the revelatio mechaism does have the origial equilibrium of iterest however, i the case of idirect mechaisms, eve if the idirect mechaism had a uique equilibrium, the revelatio mechaism ca also have ew, bad equilibria So what is the revelatio priciple good for? recogitio that truthfuless is ot a restrictive assumptio for aalysis purposes, we ca cosider oly truthful mechaisms, ad be assured that such a mechaism exists recogitio that idirect mechaisms ca t do (iheretly) better tha direct mechaisms Pigzhog Tag Mechaism Desig With Urestricted Prefereces, Slide 3

16 Revelatio Priciple Impossibility Impossibility Result Theorem (Gibbard-Satterthwaite) Cosider ay social choice fuctio C of N ad O. If: O 3 (there are at least three outcomes); 2 C is oto; that is, for every o 2 O there is a preferece profile > such that C(>) =o; ad 3 C is implemetable i domiat-strategy (= truthful i domiat-strategy = strategy-proof). the C is dictatorial. Pigzhog Tag Mechaism Desig With Urestricted Prefereces, Slide 4

17 Revelatio Priciple Impossibility What does this mea? We should be discouraged about the possibility of implemetig arbitrary social-choice fuctios i mechaisms. However, i practice we ca circumvet the Gibbard-Satterthwaite theorem i two ways: use a weaker form of implemetatio ote: the result oly holds for domiat strategy implemetatio, ot e.g., Bayes-Nash implemetatio relax the implicit assumptio that agets are allowed to hold arbitrary prefereces Pigzhog Tag Mechaism Desig With Urestricted Prefereces, Slide 5

18 Quasiliear prefereces Quasiliear Mechaisms Properties Quasiliear prefereces restricted prefereces Defiitio (Quasiliear prefereces) Agets have quasiliear prefereces i a -player Bayesia game whe the set of outcomes is O = X R for a fiite set X, ad the utility of a aget i give joit type is give by u i (o, ) =u i (x, ) f i (p i ), where o =(x, p) is a elemet of O, u i : X 7! R is a arbitrary fuctio ad f i : R 7! R is a strictly icreasig. Questio: why is Quasiliear Pref. restricted? Pigzhog Tag Quasiliear Utility, Slide

19 Quasiliear prefereces Quasiliear Mechaisms Properties Quasiliear utility u i (o, ) =u i (x, ) f i (p i ) We split the mechaism ito a choice/allocatio rule ad a paymet rule: x 2 X is a discrete, o-moetary outcome p i 2 R is a moetary paymet (possibly egative) that aget i must make to the mechaism Implicatios: Pigzhog Tag Quasiliear Utility, Slide 2

20 Quasiliear prefereces Quasiliear Mechaisms Properties Quasiliear utility u i (o, ) =u i (x, ) f i (p i ) We split the mechaism ito a choice/allocatio rule ad a paymet rule: x 2 X is a discrete, o-moetary outcome p i 2 R is a moetary paymet (possibly egative) that aget i must make to the mechaism Implicatios: u i (x, ) is ot iflueced by the amout of moey a aget has agets do t care how much others are made to pay (though they ca care about how the choice a ects others.) Pigzhog Tag Quasiliear Utility, Slide 2

21 Quasiliear prefereces Quasiliear Mechaisms Properties Quasiliear utility u i (o, ) =u i (x, ) f i (p i ) We split the mechaism ito a choice/allocatio rule ad a paymet rule: x 2 X is a discrete, o-moetary outcome p i 2 R is a moetary paymet (possibly egative) that aget i must make to the mechaism Implicatios: u i (x, ) is ot iflueced by the amout of moey a aget has agets do t care how much others are made to pay (though they ca care about how the choice a ects others.) What is f i (p i )? Pigzhog Tag Quasiliear Utility, Slide 2

22 Quasiliear prefereces Quasiliear Mechaisms Properties Quasiliear Mechaism Defiitio (Quasiliear mechaism) A mechaism i the quasiliear settig (for a Bayesia game settig (N,O = X R,,p,u)) isatriple(a, x, p), where A = A A,whereA i is the set of actios available to aget i 2 N, x : A 7! (X) maps each actio profile to a distributio over choices, ad p : A 7! R maps each actio profile to a paymet for each aget. Pigzhog Tag Quasiliear Utility, Slide 3

23 Quasiliear prefereces Quasiliear Mechaisms Properties Further simplificatios: Direct Quasiliear Mechaism Defiitio (Direct quasiliear mechaism) A direct quasiliear mechaism (for a eviromet (N,O = X R,,p,u)) isapair(x, p). Itdefiesastadard mechaism i the quasiliear settig, where for each i, A i = i. Defiitio (Coditioal utility idepedece) A Bayesia game exhibits coditioal utility idepedece if for all agets i 2 N, for all outcomes o 2 O ad for all pairs of joit types ad 0 2 for which i = 0 i, it holds that u i(o, ) =u i (o, 0 ). I other words, u i oly depeds o i. Pigzhog Tag Quasiliear Utility, Slide 4

24 Quasiliear prefereces Quasiliear Mechaisms Properties Quasiliear Mechaisms with Coditioal Utility Idepedece Give coditioal utility idepedece, we ca write i s utility fuctio as u i (o, i ) it does ot deped o the other agets types A aget s valuatio for choice x 2 X: v i (x) =u i (x, i ) the maximum amout i would be willig to pay to get x Thik of v i as i s type from ow o Alterate defiitio of direct mechaism: ask agets i to declare v i (x) for each x 2 X Defie ˆv i as the valuatio that aget i declares to such a direct mechaism may be di eret from his true valuatio v i Also defie the tuples ˆv, ˆv i Pigzhog Tag Quasiliear Utility, Slide 5

25 Quasiliear prefereces Quasiliear Mechaisms Properties Truthfuless Defiitio (Truthfuless) Aquasiliearmechaismistruthful if it is direct ad 8i8v i, aget i s equilibrium strategy is to adopt the strategy ˆv i = v i. Our defiitio before, adapted for the quasiliear settig Pigzhog Tag Quasiliear Utility, Slide 6

26 Quasiliear prefereces Quasiliear Mechaisms Properties E ciecy Defiitio (E ciecy) Aquasiliearmechaismisstrictly Pareto e ciet, orjust e ciet, if i equilibrium it selects a choice x such that 8v8x 0, X X v i (x) v i (x 0 ). i i A e ciet mechaism selects the choice that maximizes the sum of agets utilities, disregardig moetary paymets. How is this related to the Pareto e ciecy from GT? Pigzhog Tag Quasiliear Utility, Slide 7

27 Quasiliear prefereces Quasiliear Mechaisms Properties E ciecy Defiitio (E ciecy) Aquasiliearmechaismisstrictly Pareto e ciet, orjust e ciet, if i equilibrium it selects a choice x such that 8v8x 0, X X v i (x) v i (x 0 ). i i A e ciet mechaism selects the choice that maximizes the sum of agets utilities, disregardig moetary paymets. How is this related to the Pareto e ciecy from GT? if we iclude the mechaism as a aget, all Pareto-e ciet outcomes ivolve the same choice (ad di eret paymets) ay outcome ivolvig aother choice is Pareto-domiated: some agets could make a side-paymet to others such that all would prefer the swap Pigzhog Tag Quasiliear Utility, Slide 7

28 Quasiliear prefereces Quasiliear Mechaisms Properties E ciecy Defiitio (E ciecy) Aquasiliearmechaismisstrictly Pareto e ciet, orjust e ciet, if i equilibrium it selects a choice x such that 8v8x 0, X X v i (x) v i (x 0 ). i i Called ecoomic e ciecy to distiguish from other (e.g., computatioal) otios Also called social-welfare maximizatio Note: defied i terms of true (ot declared) valuatios. Pigzhog Tag Quasiliear Utility, Slide 7

29 Quasiliear prefereces Quasiliear Mechaisms Properties Budget Balace Defiitio (Budget balace) A quasiliear mechaism is budget balaced whe 8v, X i p i (s(v)) = 0, where s is the equilibrium strategy profile. regardless of the agets types, the mechaism collects ad disburses the same amout of moey from ad to the agets Pigzhog Tag Quasiliear Utility, Slide 8

30 Quasiliear prefereces Quasiliear Mechaisms Properties Budget Balace Defiitio (Budget balace) A quasiliear mechaism is budget balaced whe 8v, X i p i (s(v)) = 0, where s is the equilibrium strategy profile. regardless of the agets types, the mechaism collects ad disburses the same amout of moey from ad to the agets relaxed versio: weak budget balace: 8v, X i p i (s(v)) 0 the mechaism ever takes a loss, but it may make a profit Pigzhog Tag Quasiliear Utility, Slide 8

31 Quasiliear prefereces Quasiliear Mechaisms Properties Budget Balace Defiitio (Budget balace) A quasiliear mechaism is budget balaced whe 8v, X i p i (s(v)) = 0, where s is the equilibrium strategy profile. regardless of the agets types, the mechaism collects ad disburses the same amout of moey from ad to the agets Budget balace ca be required to hold ex ate: X E v p i (s(v)) = 0 i the mechaism must break eve or make a profit oly o expectatio Pigzhog Tag Quasiliear Utility, Slide 8

32 Quasiliear prefereces Quasiliear Mechaisms Properties Idividual-Ratioality Defiitio (Ex iterim idividual ratioality) A mechaism is ex iterim idividual ratioal whe 8i8v i, E v i v i v i (x (s i (v i ),s i (v i ))) p i (s i (v i ),s i (v i )) 0, where s is the equilibrium strategy profile. o aget loses by participatig i the mechaism. ex iterim because it holds for every possible valuatio for aget i, but averages over the possible valuatios of the other agets. Pigzhog Tag Quasiliear Utility, Slide 9

33 Quasiliear prefereces Quasiliear Mechaisms Properties Idividual-Ratioality Defiitio (Ex iterim idividual ratioality) A mechaism is ex iterim idividual ratioal whe 8i8v i, E v i v i v i (x (s i (v i ),s i (v i ))) p i (s i (v i ),s i (v i )) 0, where s is the equilibrium strategy profile. o aget loses by participatig i the mechaism. ex iterim because it holds for every possible valuatio for aget i, but averages over the possible valuatios of the other agets. Defiitio (Ex post idividual ratioality) Amechaismisex post idividual ratioal whe 8i8v, v i (x (s(v))) p i (s(v)) 0, wheres is the equilibrium strategy profile. Pigzhog Tag Quasiliear Utility, Slide 9

34 Fially, a positive result! Recall that i the quasiliear utility settig, a mechaism ca be defied as a choice rule ad a paymet rule. The Groves mechaism is a mechaism that satisfies: truthfuless i domiat strategies e ciecy I geeral it s ot: budget balaced idividual-ratioal...though we ll see later that we ca recover these properties. Pigzhog Tag Groves Mechaism, Slide

35 The Groves Mechaism Defiitio (Groves mechaism) The Groves mechaism is a direct quasiliear mechaism (x, p), where X x (ˆv) = arg max ˆv i (x) x i X p i (ˆv) =h i (ˆv i ) ˆv j (x (ˆv)) j6=i Pigzhog Tag Groves Mechaism, Slide 2

36 The Groves Mechaism x (ˆv) = arg max x p i (ˆv) =h i (ˆv i ) X ˆv i (x) i X ˆv j (x (ˆv)) j6=i The choice rule should ot come as a surprise (why ot?) Pigzhog Tag Groves Mechaism, Slide 3

37 The Groves Mechaism x (ˆv) = arg max x p i (ˆv) =h i (ˆv i ) X ˆv i (x) i X ˆv j (x (ˆv)) j6=i The choice rule should ot come as a surprise (why ot?) Give the mechaism is both truthful ad e ciet: these properties etail the give choice rule. Pigzhog Tag Groves Mechaism, Slide 3

38 The Groves Mechaism x (ˆv) = arg max x p i (ˆv) =h i (ˆv i ) X ˆv i (x) i X ˆv j (x (ˆv)) j6=i The choice rule should ot come as a surprise (why ot?) Give the mechaism is both truthful ad e ciet: these properties etail the give choice rule. So what s goig o with the paymet rule? the aget i must pay some amout h i (ˆv i ) that does t deped o his ow declared valuatio the aget i is paid P j6=i ˆv j(x (ˆv)), the sum of the others valuatios for the chose outcome Pigzhog Tag Groves Mechaism, Slide 3

39 Groves Truthfuless Theorem Truth tellig is a domiat strategy uder the Groves mechaism. Cosider a situatio where every aget j other tha i follows some arbitrary strategy ˆv j.cosiderageti s problem of choosig the best strategy ˆv i.asa shorthad, we will write ˆv =(ˆv i, ˆv i ). The best strategy for i is oe that solves max ˆv i v i (x (ˆv)) p(ˆv) Substitutig i the paymet fuctio from the Groves mechaism, we have 0 max ˆv i v i (x (ˆv)) h i (ˆv i )+ X j6=i ˆv j (x (ˆv)) A Sice h i (ˆv i ) does ot deped o ˆv i,itissu 0 ciet to solve max ˆv i v i (x (ˆv)) + X j6=i ˆv j (x (ˆv)) A. Pigzhog Tag Groves Mechaism, Slide 4

40 Groves Truthfuless max ˆv i 0 v i (x (ˆv)) + X j6=i ˆv j (x (ˆv)) A. The oly way the declaratio ˆv i iflueces this maximizatio is through the choice of x. If possible, i would like to pick a declaratio ˆv i that will lead the mechaism to pick a x 2 X which solves 0 max x v i (x)+ X j6=i ˆv j (x) A. () Uder the Groves mechaism,! X x (ˆv) = arg max ˆv i (x) x i = arg max x 0 ˆv i (x)+ X j6=i ˆv j (x) A. The Groves mechaism will choose x i a way that solves the maximizatio problem i Equatio () whe i declares ˆv i = v i.becausethisargumetdoes ot deped i ay way o the declaratios of the other agets, truth-tellig is a domiat strategy for aget i. Pigzhog Tag Groves Mechaism, Slide 5

41 Proof ituitio exteralities are iteralized agets may be able to chage the outcome to aother oe that they prefer, by chagig their declaratio however, their utility does t just deped o the outcome it also depeds o their paymet sice they get paid the (reported) utility of all the other agets uder the chose allocatio, they ow have a iterest i maximizig everyoe s utility rather tha just their ow i geeral, DS truthful mechaisms have the property that a aget s paymet does t deped o the amout of his declaratio, but oly o the other agets declaratios the aget s declaratio is used oly to choose the outcome, ad to set other agets paymets Pigzhog Tag Groves Mechaism, Slide 6

42 Groves Uiqueess Theorem (Gree La ot) A e ciet social choice fuctio C : R X! X R ca be implemeted i domiat strategies for agets P with urestricted quasiliear utilities oly if p i (v) =h(v i ) j6=i v j(x (v)). it turs out that the same result also holds for the broader class of Bayes Nash icetive-compatible e ciet mechaisms. Pigzhog Tag Groves Mechaism, Slide 7

43 VCG mechaism Defiitio (Clarke tax) The Clarke tax sets the h i term i a Groves mechaism as h i (ˆv i )= X j6=i ˆv j (x (ˆv i )). Defiitio (Vickrey-Clarke-Groves (VCG) mechaism) The Vickrey-Clarke-Groves mechaism is a direct quasiliear mechaism (x, p), where x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X ˆv j (x (ˆv)) j6=i Pigzhog Tag VCG, Slide

44 VCG discussio x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X j6=i ˆv j (x (ˆv)) You get paid everyoe s utility uder the allocatio that is actually chose except your ow, but you get that directly as utility The you get charged everyoe s utility i the world where you do t participate Thus you pay your social cost Pigzhog Tag VCG, Slide 2

45 VCG discussio Questios: who pays 0? x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X j6=i ˆv j (x (ˆv)) Pigzhog Tag VCG, Slide 3

46 VCG discussio x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X j6=i ˆv j (x (ˆv)) Questios: who pays 0? agets who do t a ect the outcome Pigzhog Tag VCG, Slide 3

47 VCG discussio x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X j6=i ˆv j (x (ˆv)) Questios: who pays 0? agets who do t a ect the outcome who pays more tha 0? Pigzhog Tag VCG, Slide 3

48 VCG discussio x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X j6=i ˆv j (x (ˆv)) Questios: who pays 0? agets who do t a ect the outcome who pays more tha 0? (pivotal) agets who make thigs worse for others by existig Pigzhog Tag VCG, Slide 3

49 VCG discussio x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X j6=i ˆv j (x (ˆv)) Questios: who pays 0? agets who do t a ect the outcome who pays more tha 0? (pivotal) agets who make thigs worse for others by existig who gets paid? Pigzhog Tag VCG, Slide 3

50 VCG discussio x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X j6=i ˆv j (x (ˆv)) Questios: who pays 0? agets who do t a ect the outcome who pays more tha 0? (pivotal) agets who make thigs worse for others by existig who gets paid? (pivotal) agets who make thigs better for others by existig Pigzhog Tag VCG, Slide 3

51 VCG properties x (ˆv) = arg max x p i (ˆv) = X j6=i X ˆv i (x) i ˆv j (x (ˆv i )) X j6=i ˆv j (x (ˆv)) Because oly pivotal agets have to pay, VCG is also called the pivot mechaism It s domiat-strategy truthful, because it s a Groves mechaism Pigzhog Tag VCG, Slide 4

52 Selfish routig example A F C E B D R - R - R What outcome will be selected by x? Pigzhog Tag VCG, Slide 5

53 Selfish routig example A F C E B D R - R - R What outcome will be selected by x? path ABEF. Pigzhog Tag VCG, Slide 5

54 Selfish routig example A F C E B D R - R - R What outcome will be selected by x? path ABEF. How much will AC have to pay? Pigzhog Tag VCG, Slide 5

55 Selfish routig example 3 A 2 R B C R - D 2 R 5 E F What outcome will be selected by x? path ABEF. How much will AC have to pay? The shortest path takig his declaratio ito accout has a legth of 5, ad imposes a cost of 5 o agets other tha him (because it does ot ivolve him). Likewise, the shortest path without AC s declaratio also has a legth of 5. Thus, his paymet p AC =( 5) ( 5) = 0. This is what we expect, sice AC is ot pivotal. Likewise, BD, CE, CF ad DF will all pay zero. Pigzhog Tag VCG, Slide 5

56 Selfish routig example A F C E B D R - R - R How much will AB pay? Pigzhog Tag VCG, Slide 6

57 Selfish routig example 3 A 2 R B C R - D 2 R 5 E F How much will AB pay? The shortest path takig AB s declaratio ito accout has a legth of 5, ad imposes a cost of 2 o other agets. The shortest path without AB is ACEF, which has a cost of 6. Thus p AB =( 6) ( 2) = 4. Pigzhog Tag VCG, Slide 6

58 Selfish routig example A F C E B D R - R - R How much will BE pay? Pigzhog Tag VCG, Slide 7

59 Selfish routig example A F C E B D R - R - R How much will BE pay? p BE =( 6) ( 4) = 2. Pigzhog Tag VCG, Slide 7

60 Selfish routig example A F C E B D R - R - R How much will BE pay? p BE =( 6) ( 4) = 2. How much will EF pay? Pigzhog Tag VCG, Slide 7

61 Selfish routig example A F C E B D R - R - R How much will BE pay? p BE =( 6) ( 4) = 2. How much will EF pay? p EF =( 7) ( 4) = 3. Pigzhog Tag VCG, Slide 7

62 Selfish routig example A F C E B D R - R - R How much will BE pay? p BE =( 6) ( 4) = 2. How much will EF pay? p EF =( 7) ( 4) = 3. EF ad BE have the same costs but are paid di eret amouts. Why? Pigzhog Tag VCG, Slide 7

63 Selfish routig example 3 A 2 R B C R - D 2 R 5 E F How much will BE pay? p BE =( 6) ( 4) = 2. How much will EF pay? p EF =( 7) ( 4) = 3. EF ad BE have the same costs but are paid di eret amouts. Why? EF has more market power: for the other agets, the situatio without EF is worse tha the situatio without BE. Pigzhog Tag VCG, Slide 7

Posted-Price, Sealed-Bid Auctions

Posted-Price, Sealed-Bid Auctions Posted-Price, Sealed-Bid Auctios Professors Greewald ad Oyakawa 207-02-08 We itroduce the posted-price, sealed-bid auctio. This auctio format itroduces the idea of approximatios. We describe how well this

More information

Economics 2450A: Public Economics Section 3: Mirrlees Taxation

Economics 2450A: Public Economics Section 3: Mirrlees Taxation Ecoomics 2450A: Public Ecoomics Sectio 3: Mirrlees Taxatio Matteo Paradisi September 26, 2016 I today s sectio we will solve the Mirrlees tax problem. We will ad derive optimal taxes itroducig the cocept

More information

1 Games in Normal Form (Strategic Form)

1 Games in Normal Form (Strategic Form) 1 Games i Normal Form Strategic Form) A Game i Normal strategic) Form cosists of three compoets: 1 A set of players For each player, a set of strategies called actios i textbook) The iterpretatio is that

More information

Public Good Provision with Constitutional Constraint

Public Good Provision with Constitutional Constraint Public Good Provisio with Costitutioal Costrait Kag Rog School of Ecoomics, Shaghai Uiversity of Fiace ad Ecoomics, Shaghai 200433, Chia Key Laboratory of Mathematical Ecoomics (SUFE), Miistry of Educatio,

More information

Public Good Provision with Constitutional Constraint

Public Good Provision with Constitutional Constraint Public Good Provisio with Costitutioal Costrait Kag Rog 1 School of Ecoomics, Shaghai Uiversity of Fiace ad Ecoomics, Shaghai 200433, Chia Key Laboratory of Mathematical Ecoomics (SUFE), Miistry of Educatio,

More information

Mechanism Design for Locating a Facility under Partial Information

Mechanism Design for Locating a Facility under Partial Information Mechaism Desig for Locatig a Facility uder Partial Iformatio Vijay Meo Kate Larso Abstract We study the classic mechaism desig problem of locatig a public facility o a real lie. However, i cotrast to previous

More information

Approximating Optimal Combinatorial Auctions for Complements Using Restricted Welfare Maximization

Approximating Optimal Combinatorial Auctions for Complements Using Restricted Welfare Maximization Approximatig Optimal Combiatorial Auctios for Complemets Usig Restricted Welfare Maximizatio Pigzhog Tag 1,2 ad Tuomas Sadholm 1 keshi@cs.cmu.edu, sadholm@cs.cmu.edu 1 Computer Sciece Departmet, Caregie

More information

Economics 241B Relation to Method of Moments and Maximum Likelihood OLSE as a Maximum Likelihood Estimator

Economics 241B Relation to Method of Moments and Maximum Likelihood OLSE as a Maximum Likelihood Estimator Ecoomics 24B Relatio to Method of Momets ad Maximum Likelihood OLSE as a Maximum Likelihood Estimator Uder Assumptio 5 we have speci ed the distributio of the error, so we ca estimate the model parameters

More information

Economics 2450A: Public Economics Section 3-4: Mirrlees Taxation

Economics 2450A: Public Economics Section 3-4: Mirrlees Taxation Ecoomics 2450A: Public Ecoomics Sectio 3-4: Mirrlees Taxatio Matteo Paradisi October 3, 206 I today s sectio we will solve the Mirrlees tax problem. We will ad derive optimal taxes itroducig the cocept

More information

Chapter 3. Strong convergence. 3.1 Definition of almost sure convergence

Chapter 3. Strong convergence. 3.1 Definition of almost sure convergence Chapter 3 Strog covergece As poited out i the Chapter 2, there are multiple ways to defie the otio of covergece of a sequece of radom variables. That chapter defied covergece i probability, covergece i

More information

Revenue Equivalence Theorem

Revenue Equivalence Theorem Reveue Equivalece Theorem Felix Muoz-Garcia Advaced Microecoomics II Washigto State Uiversity So far, several di eret auctio types have bee cosidered. The questio remais: How does the expected reveue for

More information

Sequences A sequence of numbers is a function whose domain is the positive integers. We can see that the sequence

Sequences A sequence of numbers is a function whose domain is the positive integers. We can see that the sequence Sequeces A sequece of umbers is a fuctio whose domai is the positive itegers. We ca see that the sequece 1, 1, 2, 2, 3, 3,... is a fuctio from the positive itegers whe we write the first sequece elemet

More information

ECO 312 Fall 2013 Chris Sims LIKELIHOOD, POSTERIORS, DIAGNOSING NON-NORMALITY

ECO 312 Fall 2013 Chris Sims LIKELIHOOD, POSTERIORS, DIAGNOSING NON-NORMALITY ECO 312 Fall 2013 Chris Sims LIKELIHOOD, POSTERIORS, DIAGNOSING NON-NORMALITY (1) A distributio that allows asymmetry differet probabilities for egative ad positive outliers is the asymmetric double expoetial,

More information

A sequence of numbers is a function whose domain is the positive integers. We can see that the sequence

A sequence of numbers is a function whose domain is the positive integers. We can see that the sequence Sequeces A sequece of umbers is a fuctio whose domai is the positive itegers. We ca see that the sequece,, 2, 2, 3, 3,... is a fuctio from the positive itegers whe we write the first sequece elemet as

More information

Lecture 19: Convergence

Lecture 19: Convergence Lecture 19: Covergece Asymptotic approach I statistical aalysis or iferece, a key to the success of fidig a good procedure is beig able to fid some momets ad/or distributios of various statistics. I may

More information

Understanding Samples

Understanding Samples 1 Will Moroe CS 109 Samplig ad Bootstrappig Lecture Notes #17 August 2, 2017 Based o a hadout by Chris Piech I this chapter we are goig to talk about statistics calculated o samples from a populatio. We

More information

Mathematical Induction

Mathematical Induction Mathematical Iductio Itroductio Mathematical iductio, or just iductio, is a proof techique. Suppose that for every atural umber, P() is a statemet. We wish to show that all statemets P() are true. I a

More information

Bertrand s Postulate

Bertrand s Postulate Bertrad s Postulate Lola Thompso Ross Program July 3, 2009 Lola Thompso (Ross Program Bertrad s Postulate July 3, 2009 1 / 33 Bertrad s Postulate I ve said it oce ad I ll say it agai: There s always a

More information

Product measures, Tonelli s and Fubini s theorems For use in MAT3400/4400, autumn 2014 Nadia S. Larsen. Version of 13 October 2014.

Product measures, Tonelli s and Fubini s theorems For use in MAT3400/4400, autumn 2014 Nadia S. Larsen. Version of 13 October 2014. Product measures, Toelli s ad Fubii s theorems For use i MAT3400/4400, autum 2014 Nadia S. Larse Versio of 13 October 2014. 1. Costructio of the product measure The purpose of these otes is to preset the

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

Design and Analysis of Algorithms

Design and Analysis of Algorithms Desig ad Aalysis of Algorithms Probabilistic aalysis ad Radomized algorithms Referece: CLRS Chapter 5 Topics: Hirig problem Idicatio radom variables Radomized algorithms Huo Hogwei 1 The hirig problem

More information

Statistical Pattern Recognition

Statistical Pattern Recognition Statistical Patter Recogitio Classificatio: No-Parametric Modelig Hamid R. Rabiee Jafar Muhammadi Sprig 2014 http://ce.sharif.edu/courses/92-93/2/ce725-2/ Ageda Parametric Modelig No-Parametric Modelig

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

Induction: Solutions

Induction: Solutions Writig Proofs Misha Lavrov Iductio: Solutios Wester PA ARML Practice March 6, 206. Prove that a 2 2 chessboard with ay oe square removed ca always be covered by shaped tiles. Solutio : We iduct o. For

More information

Definition 4.2. (a) A sequence {x n } in a Banach space X is a basis for X if. unique scalars a n (x) such that x = n. a n (x) x n. (4.

Definition 4.2. (a) A sequence {x n } in a Banach space X is a basis for X if. unique scalars a n (x) such that x = n. a n (x) x n. (4. 4. BASES I BAACH SPACES 39 4. BASES I BAACH SPACES Sice a Baach space X is a vector space, it must possess a Hamel, or vector space, basis, i.e., a subset {x γ } γ Γ whose fiite liear spa is all of X ad

More information

Notes for Lecture 5. 1 Grover Search. 1.1 The Setting. 1.2 Motivation. Lecture 5 (September 26, 2018)

Notes for Lecture 5. 1 Grover Search. 1.1 The Setting. 1.2 Motivation. Lecture 5 (September 26, 2018) COS 597A: Quatum Cryptography Lecture 5 (September 6, 08) Lecturer: Mark Zhadry Priceto Uiversity Scribe: Fermi Ma Notes for Lecture 5 Today we ll move o from the slightly cotrived applicatios of quatum

More information

Distribution of Random Samples & Limit theorems

Distribution of Random Samples & Limit theorems STAT/MATH 395 A - PROBABILITY II UW Witer Quarter 2017 Néhémy Lim Distributio of Radom Samples & Limit theorems 1 Distributio of i.i.d. Samples Motivatig example. Assume that the goal of a study is to

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit Theorems Throughout this sectio we will assume a probability space (, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

Refinements. New notes follow on next page. 1 of 8. Equilibrium path. Belief needed

Refinements. New notes follow on next page. 1 of 8. Equilibrium path. Belief needed efiemets Itro from ECO 744 Notes efiemets i Strategic Form: Elimiate Weakly Domiated Strategies - throwig out strictly domiated strategies (eve iteratively ever elimiates a Nash equilibrium (because Nash

More information

An Exploration in the Theory of Optimum Income Taxation

An Exploration in the Theory of Optimum Income Taxation A Exploratio i the Theory of Optimum Icome Taxatio J. A. Mirrlees, 1971 A Exploratio i the Theory of Optimum Icome Taxatio p. 1/12 Objectives ad Results Characterize the optimal icome tax schedule whe

More information

Vickrey Auction. Mechanism Design. Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory 2013 Mechanism Design

Vickrey Auction. Mechanism Design. Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory 2013 Mechanism Design Algorithmic Game Theory Vickrey Auction Vickrey-Clarke-Groves Mechanisms Mechanisms with Money Player preferences are quantifiable. Common currency enables utility transfer between players. Preference

More information

Lecture 10 October Minimaxity and least favorable prior sequences

Lecture 10 October Minimaxity and least favorable prior sequences STATS 300A: Theory of Statistics Fall 205 Lecture 0 October 22 Lecturer: Lester Mackey Scribe: Brya He, Rahul Makhijai Warig: These otes may cotai factual ad/or typographic errors. 0. Miimaxity ad least

More information

Vickrey Auction VCG Characterization. Mechanism Design. Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory 2013 Mechanism Design

Vickrey Auction VCG Characterization. Mechanism Design. Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory 2013 Mechanism Design Algorithmic Game Theory Vickrey Auction Vickrey-Clarke-Groves Mechanisms Characterization of IC Mechanisms Mechanisms with Money Player preferences are quantifiable. Common currency enables utility transfer

More information

Notes 19 : Martingale CLT

Notes 19 : Martingale CLT Notes 9 : Martigale CLT Math 733-734: Theory of Probability Lecturer: Sebastie Roch Refereces: [Bil95, Chapter 35], [Roc, Chapter 3]. Sice we have ot ecoutered weak covergece i some time, we first recall

More information

1 of 7 7/16/2009 6:06 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 6. Order Statistics Defiitios Suppose agai that we have a basic radom experimet, ad that X is a real-valued radom variable

More information

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 5

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 5 CS434a/54a: Patter Recogitio Prof. Olga Veksler Lecture 5 Today Itroductio to parameter estimatio Two methods for parameter estimatio Maimum Likelihood Estimatio Bayesia Estimatio Itroducto Bayesia Decisio

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

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS MASSACHUSTTS INSTITUT OF TCHNOLOGY 6.436J/5.085J Fall 2008 Lecture 9 /7/2008 LAWS OF LARG NUMBRS II Cotets. The strog law of large umbers 2. The Cheroff boud TH STRONG LAW OF LARG NUMBRS While the weak

More information

( ) = p and P( i = b) = q.

( ) = p and P( i = b) = q. MATH 540 Radom Walks Part 1 A radom walk X is special stochastic process that measures the height (or value) of a particle that radomly moves upward or dowward certai fixed amouts o each uit icremet of

More information

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f.

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f. Lecture 5 Let us give oe more example of MLE. Example 3. The uiform distributio U[0, ] o the iterval [0, ] has p.d.f. { 1 f(x =, 0 x, 0, otherwise The likelihood fuctio ϕ( = f(x i = 1 I(X 1,..., X [0,

More information

A Simple Budget-balanced Mechanism

A Simple Budget-balanced Mechanism A Simple Budget-balaced Mechaism Debasis Mishra ad Tridib Sharma November 10, 2016 Abstract I the private values sigle object auctio model, we costruct a satisfactory mechaism - a symmetric, domiat strategy

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

Statistics 511 Additional Materials

Statistics 511 Additional Materials Cofidece Itervals o mu Statistics 511 Additioal Materials This topic officially moves us from probability to statistics. We begi to discuss makig ifereces about the populatio. Oe way to differetiate probability

More information

Hashing and Amortization

Hashing and Amortization Lecture Hashig ad Amortizatio Supplemetal readig i CLRS: Chapter ; Chapter 7 itro; Sectio 7.. Arrays ad Hashig Arrays are very useful. The items i a array are statically addressed, so that isertig, deletig,

More information

Once we have a sequence of numbers, the next thing to do is to sum them up. Given a sequence (a n ) n=1

Once we have a sequence of numbers, the next thing to do is to sum them up. Given a sequence (a n ) n=1 . Ifiite Series Oce we have a sequece of umbers, the ext thig to do is to sum them up. Give a sequece a be a sequece: ca we give a sesible meaig to the followig expressio? a = a a a a While summig ifiitely

More information

Exponential Families and Bayesian Inference

Exponential Families and Bayesian Inference Computer Visio Expoetial Families ad Bayesia Iferece Lecture Expoetial Families A expoetial family of distributios is a d-parameter family f(x; havig the followig form: f(x; = h(xe g(t T (x B(, (. where

More information

Principle Of Superposition

Principle Of Superposition ecture 5: PREIMINRY CONCEP O RUCUR NYI Priciple Of uperpositio Mathematically, the priciple of superpositio is stated as ( a ) G( a ) G( ) G a a or for a liear structural system, the respose at a give

More information

John Riley 30 August 2016

John Riley 30 August 2016 Joh Riley 3 August 6 Basic mathematics of ecoomic models Fuctios ad derivatives Limit of a fuctio Cotiuity 3 Level ad superlevel sets 3 4 Cost fuctio ad margial cost 4 5 Derivative of a fuctio 5 6 Higher

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

(A sequence also can be thought of as the list of function values attained for a function f :ℵ X, where f (n) = x n for n 1.) x 1 x N +k x N +4 x 3

(A sequence also can be thought of as the list of function values attained for a function f :ℵ X, where f (n) = x n for n 1.) x 1 x N +k x N +4 x 3 MATH 337 Sequeces Dr. Neal, WKU Let X be a metric space with distace fuctio d. We shall defie the geeral cocept of sequece ad limit i a metric space, the apply the results i particular to some special

More information

CS / MCS 401 Homework 3 grader solutions

CS / MCS 401 Homework 3 grader solutions CS / MCS 401 Homework 3 grader solutios assigmet due July 6, 016 writte by Jāis Lazovskis maximum poits: 33 Some questios from CLRS. Questios marked with a asterisk were ot graded. 1 Use the defiitio of

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit theorems Throughout this sectio we will assume a probability space (Ω, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1 EECS564 Estimatio, Filterig, ad Detectio Hwk 2 Sols. Witer 25 4. Let Z be a sigle observatio havig desity fuctio where. p (z) = (2z + ), z (a) Assumig that is a oradom parameter, fid ad plot the maximum

More information

Shannon s noiseless coding theorem

Shannon s noiseless coding theorem 18.310 lecture otes May 4, 2015 Shao s oiseless codig theorem Lecturer: Michel Goemas I these otes we discuss Shao s oiseless codig theorem, which is oe of the foudig results of the field of iformatio

More information

b i u x i U a i j u x i u x j

b i u x i U a i j u x i u x j M ath 5 2 7 Fall 2 0 0 9 L ecture 1 9 N ov. 1 6, 2 0 0 9 ) S ecod- Order Elliptic Equatios: Weak S olutios 1. Defiitios. I this ad the followig two lectures we will study the boudary value problem Here

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

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

Chapter 6 Principles of Data Reduction

Chapter 6 Principles of Data Reduction Chapter 6 for BST 695: Special Topics i Statistical Theory. Kui Zhag, 0 Chapter 6 Priciples of Data Reductio Sectio 6. Itroductio Goal: To summarize or reduce the data X, X,, X to get iformatio about a

More information

Empirical Process Theory and Oracle Inequalities

Empirical Process Theory and Oracle Inequalities Stat 928: Statistical Learig Theory Lecture: 10 Empirical Process Theory ad Oracle Iequalities Istructor: Sham Kakade 1 Risk vs Risk See Lecture 0 for a discussio o termiology. 2 The Uio Boud / Boferoi

More information

1 Review and Overview

1 Review and Overview DRAFT a fial versio will be posted shortly CS229T/STATS231: Statistical Learig Theory Lecturer: Tegyu Ma Lecture #3 Scribe: Migda Qiao October 1, 2013 1 Review ad Overview I the first half of this course,

More information

The Maximum-Likelihood Decoding Performance of Error-Correcting Codes

The Maximum-Likelihood Decoding Performance of Error-Correcting Codes The Maximum-Lielihood Decodig Performace of Error-Correctig Codes Hery D. Pfister ECE Departmet Texas A&M Uiversity August 27th, 2007 (rev. 0) November 2st, 203 (rev. ) Performace of Codes. Notatio X,

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

Lecture 7: October 18, 2017

Lecture 7: October 18, 2017 Iformatio ad Codig Theory Autum 207 Lecturer: Madhur Tulsiai Lecture 7: October 8, 207 Biary hypothesis testig I this lecture, we apply the tools developed i the past few lectures to uderstad the problem

More information

REGRESSION WITH QUADRATIC LOSS

REGRESSION WITH QUADRATIC LOSS REGRESSION WITH QUADRATIC LOSS MAXIM RAGINSKY Regressio with quadratic loss is aother basic problem studied i statistical learig theory. We have a radom couple Z = X, Y ), where, as before, X is a R d

More information

CS284A: Representations and Algorithms in Molecular Biology

CS284A: Representations and Algorithms in Molecular Biology CS284A: Represetatios ad Algorithms i Molecular Biology Scribe Notes o Lectures 3 & 4: Motif Discovery via Eumeratio & Motif Represetatio Usig Positio Weight Matrix Joshua Gervi Based o presetatios by

More information

An Introduction to Randomized Algorithms

An Introduction to Randomized Algorithms A Itroductio to Radomized Algorithms The focus of this lecture is to study a radomized algorithm for quick sort, aalyze it usig probabilistic recurrece relatios, ad also provide more geeral tools for aalysis

More information

EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY

EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY GRADUATE DIPLOMA, 016 MODULE : Statistical Iferece Time allowed: Three hours Cadidates should aswer FIVE questios. All questios carry equal marks. The umber

More information

1 Hash tables. 1.1 Implementation

1 Hash tables. 1.1 Implementation Lecture 8 Hash Tables, Uiversal Hash Fuctios, Balls ad Bis Scribes: Luke Johsto, Moses Charikar, G. Valiat Date: Oct 18, 2017 Adapted From Virgiia Williams lecture otes 1 Hash tables A hash table is a

More information

Problem Set 2 Solutions

Problem Set 2 Solutions CS271 Radomess & Computatio, Sprig 2018 Problem Set 2 Solutios Poit totals are i the margi; the maximum total umber of poits was 52. 1. Probabilistic method for domiatig sets 6pts Pick a radom subset S

More information

Beurling Integers: Part 2

Beurling Integers: Part 2 Beurlig Itegers: Part 2 Isomorphisms Devi Platt July 11, 2015 1 Prime Factorizatio Sequeces I the last article we itroduced the Beurlig geeralized itegers, which ca be represeted as a sequece of real umbers

More information

Goodness-of-Fit Tests and Categorical Data Analysis (Devore Chapter Fourteen)

Goodness-of-Fit Tests and Categorical Data Analysis (Devore Chapter Fourteen) Goodess-of-Fit Tests ad Categorical Data Aalysis (Devore Chapter Fourtee) MATH-252-01: Probability ad Statistics II Sprig 2019 Cotets 1 Chi-Squared Tests with Kow Probabilities 1 1.1 Chi-Squared Testig................

More information

Regression, Part I. A) Correlation describes the relationship between two variables, where neither is independent or a predictor.

Regression, Part I. A) Correlation describes the relationship between two variables, where neither is independent or a predictor. Regressio, Part I I. Differece from correlatio. II. Basic idea: A) Correlatio describes the relatioship betwee two variables, where either is idepedet or a predictor. - I correlatio, it would be irrelevat

More information

Recap Social Choice Functions Fun Game Mechanism Design. Mechanism Design. Lecture 13. Mechanism Design Lecture 13, Slide 1

Recap Social Choice Functions Fun Game Mechanism Design. Mechanism Design. Lecture 13. Mechanism Design Lecture 13, Slide 1 Mechanism Design Lecture 13 Mechanism Design Lecture 13, Slide 1 Lecture Overview 1 Recap 2 Social Choice Functions 3 Fun Game 4 Mechanism Design Mechanism Design Lecture 13, Slide 2 Notation N is the

More information

Random Models. Tusheng Zhang. February 14, 2013

Random Models. Tusheng Zhang. February 14, 2013 Radom Models Tusheg Zhag February 14, 013 1 Radom Walks Let me describe the model. Radom walks are used to describe the motio of a movig particle (object). Suppose that a particle (object) moves alog the

More information

A quick activity - Central Limit Theorem and Proportions. Lecture 21: Testing Proportions. Results from the GSS. Statistics and the General Population

A quick activity - Central Limit Theorem and Proportions. Lecture 21: Testing Proportions. Results from the GSS. Statistics and the General Population A quick activity - Cetral Limit Theorem ad Proportios Lecture 21: Testig Proportios Statistics 10 Coli Rudel Flip a coi 30 times this is goig to get loud! Record the umber of heads you obtaied ad calculate

More information

Discussion of Centrality-based Capital Allocations and Bailout Funds Alter, Craig, and Raupach (2014) Alireza Tahbaz-Salehi

Discussion of Centrality-based Capital Allocations and Bailout Funds Alter, Craig, and Raupach (2014) Alireza Tahbaz-Salehi Discussio of Cetrality-based Capital Allocatios ad Bailout Fuds Alter, Craig, ad Raupach (2014) Alireza Tahbaz-Salehi Columbia Busiess School Aual Iteratioal Joural of Cetral Bakig Research Coferece Federal

More information

2 Banach spaces and Hilbert spaces

2 Banach spaces and Hilbert spaces 2 Baach spaces ad Hilbert spaces Tryig to do aalysis i the ratioal umbers is difficult for example cosider the set {x Q : x 2 2}. This set is o-empty ad bouded above but does ot have a least upper boud

More information

Stat 421-SP2012 Interval Estimation Section

Stat 421-SP2012 Interval Estimation Section Stat 41-SP01 Iterval Estimatio Sectio 11.1-11. We ow uderstad (Chapter 10) how to fid poit estimators of a ukow parameter. o However, a poit estimate does ot provide ay iformatio about the ucertaity (possible

More information

Sums, products and sequences

Sums, products and sequences Sums, products ad sequeces How to write log sums, e.g., 1+2+ (-1)+ cocisely? i=1 Sum otatio ( sum from 1 to ): i 3 = 1 + 2 + + If =3, i=1 i = 1+2+3=6. The ame ii does ot matter. Could use aother letter

More information

CHAPTER I: Vector Spaces

CHAPTER I: Vector Spaces CHAPTER I: Vector Spaces Sectio 1: Itroductio ad Examples This first chapter is largely a review of topics you probably saw i your liear algebra course. So why cover it? (1) Not everyoe remembers everythig

More information

UNIT #5. Lesson #2 Arithmetic and Geometric Sequences. Lesson #3 Summation Notation. Lesson #4 Arithmetic Series. Lesson #5 Geometric Series

UNIT #5. Lesson #2 Arithmetic and Geometric Sequences. Lesson #3 Summation Notation. Lesson #4 Arithmetic Series. Lesson #5 Geometric Series UNIT #5 SEQUENCES AND SERIES Lesso # Sequeces Lesso # Arithmetic ad Geometric Sequeces Lesso #3 Summatio Notatio Lesso #4 Arithmetic Series Lesso #5 Geometric Series Lesso #6 Mortgage Paymets COMMON CORE

More information

Sequences, Mathematical Induction, and Recursion. CSE 2353 Discrete Computational Structures Spring 2018

Sequences, Mathematical Induction, and Recursion. CSE 2353 Discrete Computational Structures Spring 2018 CSE 353 Discrete Computatioal Structures Sprig 08 Sequeces, Mathematical Iductio, ad Recursio (Chapter 5, Epp) Note: some course slides adopted from publisher-provided material Overview May mathematical

More information

Probability theory and mathematical statistics:

Probability theory and mathematical statistics: N.I. Lobachevsky State Uiversity of Nizhi Novgorod Probability theory ad mathematical statistics: Law of Total Probability. Associate Professor A.V. Zorie Law of Total Probability. 1 / 14 Theorem Let H

More information

Mechanism Design: Implementation. Game Theory Course: Jackson, Leyton-Brown & Shoham

Mechanism Design: Implementation. Game Theory Course: Jackson, Leyton-Brown & Shoham Game Theory Course: Jackson, Leyton-Brown & Shoham Bayesian Game Setting Extend the social choice setting to a new setting where agents can t be relied upon to disclose their preferences honestly Start

More information

ECE 330:541, Stochastic Signals and Systems Lecture Notes on Limit Theorems from Probability Fall 2002

ECE 330:541, Stochastic Signals and Systems Lecture Notes on Limit Theorems from Probability Fall 2002 ECE 330:541, Stochastic Sigals ad Systems Lecture Notes o Limit Theorems from robability Fall 00 I practice, there are two ways we ca costruct a ew sequece of radom variables from a old sequece of radom

More information

As stated by Laplace, Probability is common sense reduced to calculation.

As stated by Laplace, Probability is common sense reduced to calculation. Note: Hadouts DO NOT replace the book. I most cases, they oly provide a guidelie o topics ad a ituitive feel. The math details will be covered i class, so it is importat to atted class ad also you MUST

More information

Regression with quadratic loss

Regression with quadratic loss Regressio with quadratic loss Maxim Ragisky October 13, 2015 Regressio with quadratic loss is aother basic problem studied i statistical learig theory. We have a radom couple Z = X,Y, where, as before,

More information

Lecture XVI - Lifting of paths and homotopies

Lecture XVI - Lifting of paths and homotopies Lecture XVI - Liftig of paths ad homotopies I the last lecture we discussed the liftig problem ad proved that the lift if it exists is uiquely determied by its value at oe poit. I this lecture we shall

More information

Riemann Sums y = f (x)

Riemann Sums y = f (x) Riema Sums Recall that we have previously discussed the area problem I its simplest form we ca state it this way: The Area Problem Let f be a cotiuous, o-egative fuctio o the closed iterval [a, b] Fid

More information

Math 525: Lecture 5. January 18, 2018

Math 525: Lecture 5. January 18, 2018 Math 525: Lecture 5 Jauary 18, 2018 1 Series (review) Defiitio 1.1. A sequece (a ) R coverges to a poit L R (writte a L or lim a = L) if for each ǫ > 0, we ca fid N such that a L < ǫ for all N. If the

More information

CEE 522 Autumn Uncertainty Concepts for Geotechnical Engineering

CEE 522 Autumn Uncertainty Concepts for Geotechnical Engineering CEE 5 Autum 005 Ucertaity Cocepts for Geotechical Egieerig Basic Termiology Set A set is a collectio of (mutually exclusive) objects or evets. The sample space is the (collectively exhaustive) collectio

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2016 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

DEPARTMENT OF ACTUARIAL STUDIES RESEARCH PAPER SERIES

DEPARTMENT OF ACTUARIAL STUDIES RESEARCH PAPER SERIES DEPARTMENT OF ACTUARIAL STUDIES RESEARCH PAPER SERIES Icreasig ad Decreasig Auities ad Time Reversal by Jim Farmer Jim.Farmer@mq.edu.au Research Paper No. 2000/02 November 2000 Divisio of Ecoomic ad Fiacial

More information

Discrete-Time Systems, LTI Systems, and Discrete-Time Convolution

Discrete-Time Systems, LTI Systems, and Discrete-Time Convolution EEL5: Discrete-Time Sigals ad Systems. Itroductio I this set of otes, we begi our mathematical treatmet of discrete-time s. As show i Figure, a discrete-time operates or trasforms some iput sequece x [

More information

A LIMITED ARITHMETIC ON SIMPLE CONTINUED FRACTIONS - II 1. INTRODUCTION

A LIMITED ARITHMETIC ON SIMPLE CONTINUED FRACTIONS - II 1. INTRODUCTION A LIMITED ARITHMETIC ON SIMPLE CONTINUED FRACTIONS - II C. T. LONG J. H. JORDAN* Washigto State Uiversity, Pullma, Washigto 1. INTRODUCTION I the first paper [2 ] i this series, we developed certai properties

More information

Lecture 2: April 3, 2013

Lecture 2: April 3, 2013 TTIC/CMSC 350 Mathematical Toolkit Sprig 203 Madhur Tulsiai Lecture 2: April 3, 203 Scribe: Shubhedu Trivedi Coi tosses cotiued We retur to the coi tossig example from the last lecture agai: Example. Give,

More information

In this section we derive some finite-sample properties of the OLS estimator. b is an estimator of β. It is a function of the random sample data.

In this section we derive some finite-sample properties of the OLS estimator. b is an estimator of β. It is a function of the random sample data. 17 3. OLS Part III I this sectio we derive some fiite-sample properties of the OLS estimator. 3.1 The Samplig Distributio of the OLS Estimator y = Xβ + ε ; ε ~ N[0, σ 2 I ] b = (X X) 1 X y = f(y) ε is

More information

Lecture 2: Monte Carlo Simulation

Lecture 2: Monte Carlo Simulation STAT/Q SCI 43: Itroductio to Resamplig ethods Sprig 27 Istructor: Ye-Chi Che Lecture 2: ote Carlo Simulatio 2 ote Carlo Itegratio Assume we wat to evaluate the followig itegratio: e x3 dx What ca we do?

More information

Lecture 14: Graph Entropy

Lecture 14: Graph Entropy 15-859: Iformatio Theory ad Applicatios i TCS Sprig 2013 Lecture 14: Graph Etropy March 19, 2013 Lecturer: Mahdi Cheraghchi Scribe: Euiwoog Lee 1 Recap Bergma s boud o the permaet Shearer s Lemma Number

More information

Power and Type II Error

Power and Type II Error Statistical Methods I (EXST 7005) Page 57 Power ad Type II Error Sice we do't actually kow the value of the true mea (or we would't be hypothesizig somethig else), we caot kow i practice the type II error

More information