The Complexity of Early-Deciding in Unreliable Synchronous Networks Fabrice Le Fessant 1

Size: px
Start display at page:

Download "The Complexity of Early-Deciding in Unreliable Synchronous Networks Fabrice Le Fessant 1"

Transcription

1 The Complexty of Early-Decdng n Unrelable Synchronous Networks Fabrce Le Fessant 1 fabrce@lefessant.net Mcrosoft Research Lab., Cambrdge, CB3 0FB, UK Aprl 1, 2003 Techncal Report MSR-TR Mcrosoft Research Mcrosoft Corporaton One Mcrosoft Way Redmond, WA Work started at INRIA Rocquencourt, B.P. 105, Le Chesnay Cedex, France

2 1 Introducton In most dstrbuted systems, falures are unlkely to occur durng most short dstrbuted operatons, but wll stll occur from tme to tme: thus, dstrbuted algorthms are often desgned to perform very effcently n the absence of falures, whle degradng performances n the presence of falures to preserve some safety propertes. In the doman of agreement algorthms, these algorthms are called early-decdng algorthms, snce they try to reach a decson as fast as possble. Such algorthms, and ther complexty, have already been studed a lot: n [DRS86], Dolev and al. studed such algorthms for non-unform Consensus n dfferent byzantne models; n [CS00], Charron and al. studed the complexty of the Unform Consensus problem n synchronous systems wth crashes, followed by [KR01] for another proof of the same results; n [DFHR02], Delporte and al. presented another early-decdng algorthm for the Global Data Computaton (whch solves both Unform Consensus and NB-AC problems) problem n asynchronous systems wth Perfect Falure Detectors; fnally, [CL02] presented another algorthm for the k-tag problems (see secton 2) n synchronous systems wth crashes. All these papers gave lower-bounds on early-decson or exhbted algorthms to prove the reachablty of these lower-bounds. Whle for classcal algorthms, lower-bounds are expressed as a functon of t, the maxmal number of falures tolerated by the algorthm, for early-decdng algorthms, the lower-bounds are refned to take nto account f, the actual number of falures n a run. We can summarze the results presented n these papers as follows: Byzantne model: Eventually Byzantne Agreement (EBA, equvalent to Consensus) requres mn(f + 2,t + 1) rounds[drs86], and can be solved wthn these bounds wth n > max(4t,2(t + (t 1) 2 )). Smultaneous Byzantne Agreement (SBA) requres mn(n 1, t + 1) rounds [DRS86]. Synchronous model wth crashes: Unform Consensus requres f + 2 rounds f f < t 1, and f + 1 otherwse[cs00, KR01], and can be solved wthn these bounds[cs00]. k-tag (Unform Consensus and NB-AC) requres f +2 rounds f f < t 1, f +1 otherwse[cl02], and can be solved wthn these bounds[cl02]. Synchronous model wth general-omsson falures: Termnatng Relable Broadcast (TRB) requres f + 2 rounds [Ros96a, Ros96b] wth t unbounded. Asynchronous Model wth Perfect Falure-Detectors: Global Data Computaton (Unform Consensus and NB-AC) can be solved mn(f + 2, t + 1) rounds[dfhr02]. In ths paper, we study the complexty of the k-tag agreement problems [CL02] n the synchronous model, wth ether send-omsson or general-omsson falures. In both models, the lower-bounds are dfferent from the crash falures model, only for f = t 1 where f + 2 1

3 rounds are sometmes necessary. It s mportant to notce that the k-tag problems are unform problems, whch are harder to solve n these models than non-unform ones (see [Ros96a] for a smple soluton to the TRB n the general-omsson model). Our man contrbuton s to show that these lower-bounds for early-decdng can be acheved by two dfferent algorthms: the frst algorthm copes wth send-omsson falures, and solves a problem that can be vewed as a generalzaton of [DFHR02], snce the asynchronous model wth perfect falure detecton can be reduced to a synchronous model where ncorrect processes perform send-omssons only one round before ther crash. The second algorthm copes wth general-omsson falures, and requres a strct majorty of correct processes, whch s anyway the requrement of any unform agreement algorthm n that model [BN92]. The next secton presents the context of ths paper (the models and the k-tag class of agreement problems). Secton 3 and 4 respectvely descrbe the algorthms for the sendomsson and general-omsson models, together wth ther proofs. 2 Context 2.1 Model We consder a synchronous model for dstrbuted systems, as descrbed n [Lyn96] wth ether send-omssons or general-omssons. In these models, Π s a fully connected graph wth n nodes. A synchronous algorthm A s a collecton of n processes, one for each node of Π. Each process p, Π, s defned by a set of states states, a subset start states of ntal states, a message-generaton functon msgs mappng states Π to a unque (possbly null message, and a state transton functon trans mappng states and vectors ndexed by Π of messages to states. An executon of the algorthm A begns wth each process n an arbtrary start state. Then each process p, n lock-step, repeatedly performs the followng three steps: 1. Apply msgs to the current state to determne the messages to be sent and send these messages 2. Receve the messages whch are sent to p durng the prevous step 3. Apply trans to the current state and the messages that are just receved to obtan the new state. The combnaton of these three steps s called a round of A. A run of A models an executon of the entre system: t s defned by the collecton of the ntal states of processes, and the nfnte sequence of rounds of A that are executed. Wthn ths well-known synchronous model, we focus on two models of falures: Send-omsson falures: ncorrect processes may omt to send some (or all) of ther messages, and they may even crash. General-omsson falures: ncorrect processes may omt to send some messages, they may omt to receve some messages, and they may even crash. In both models, correct processes always send all ther messages n each round, and receve all the messages that were sent to them n the round. An ncorrect process whch crashes stop executng steps of the algorthm. Moreover, t can crash n the mddle of a round, maybe succeedng n sendng only a subset of the messages specfed to be sent. 2

4 In any run, f s the number of ncorrect processes, whle t s the number of ncorrect processes tolerated by the agreement algorthm consdered. 2.2 Agreement Problems In ths paper, we wll focus on the k-tag class of agreement problems [CL02], defned by the followng three condtons: Unform Agreement: No two processes (whether correct or faulty) decde dfferently. Termnaton: All correct processes eventually decde. k-valdty: 1. If at least k processes start wth 0, then 0 s the only possble decson value. 2. If all processes start wth 1 and at most k 1 falures occur, then 1 s the only possble decson value. Ths class of problems contans two well-known agreement problems: Unform Consensus(n- Tag) and Non-Blockng Atomc Commtment(1-Tag). Between them, k-tag problems can model agreement problems where a commt should only be performed f some quorum of processes do agree. The nterest of studyng ths class of problems s that our algorthms can be parameterzed by the value k, to solve any partcular problem n ths class. Contrary to the approach based on reducton between problems, there s no need n our approach to show that the results stll hold through the reducton process. 3 The Send-Omsson Model We start by presentng the lower-bounds for early-decdng n ths model: Theorem 3.1 (Lower-Bounds) Let A be a k-tag algorthm whch tolerates t process crashes. For each f, 0 f t, there exsts a run of A wth at most f crashes n whch at least one process decdes not earler then durng round f +2 f f t 1, and t+1 otherwse. Proof: The proof wll be publshed n another paper[cl03]. Ths result only dffers from the lowerbounds of the crash model[cs00] n the case f = t 1. Intutvely, n the crash model, the optmzaton n the case f = t 1 comes from the knowledge that, f a process fals durng the round t, ether ts state s lost, or another process has receved a message from t, and, snce t must be correct, wll be able to propagate ths state durng round t + 1. In the send-omsson model, the ncorrect process at round t may fal by not sendng ts state to hs neghbors, but mght stll decde, leadng to a non-unform decson f the prevous optmzaton were performed. Theorem 3.1 In Fgure 1, we present an algorthm for k-tag, that acheves these lower-bounds. The algorthm s parameterzed by a functon Φ k as n [CL02]. Before further nvestgatons, we need some defntons: Defnton 3.2 (Round quescent) For any round r > 0, for any process p, a round s sad quescent f and only f Faled (r) = Faled (r 1). 3

5 states rounds N, ntally 0 W (V { }) n, ntally (,,v,, ) where v s p s ntal value ready, ether false, true, ntally false halt, a Boolean, ntally false F aled and Rec F aled Π, ntally decson V {unknown}, ntally unknown msgs f halt and / Faled then (cond 1) send (ready,w[],faled) to all processes trans rounds := rounds + 1 Rec Faled := Faled let J be the set of ndces of processes, from whch a message has arrved let (ready j,w j,faled j ) be the message receved from p j, for each j J Faled := j J Faled j (Π\J) J := Π\Faled f halt and / Faled then (cond 1) j J, W[j] := w j j / J, W[j] := f not ready then (cond 2) W[] := Φ k (W) f F aled = Rec F aled then (cond 3) ready := true let R = {j J ready j } f R Ø then W[] := w mn(r) f rounds = mn( F aled + 2, t + 1) then decson := W[] (cond 4) Fgure 1: A k-tag Algorthm for the Send-Omsson Model 4

6 Intutvely, a round s quescent for a process f t does not detect new falures durng ths round. We also need to pont out some nterestng propertes of ths algorthm: Falures propagaton Falures are propagated between processes by messages, so that f a process has detected a falure, any process that receves some nformaton from ths process (ether drectly or ndrectly) durng the followng rounds wll also be aware of the detected falure. No ncorrect detectons Correct processes never fal to send a message. Consequently, snce process falures are detected when a message s not receved (there are no receveomssons), correct processes are never suspected, even by ncorrect processes. Now, we can prove that ths algorthm satsfes all the propertes of the k-tag problems: Theorem 3.3 (Termnaton) The algorthm satsfes Termnaton, and all correct processes decde n at most mn(f + 2,t + 1) rounds. Proof: Correct processes are never suspected by other processes, so that they never appear n the F aled sets. Then, (cond4) guarantes termnaton for them n the specfed bounds, snce F aled f. Theorem 3.3 Theorem 3.4 (Unform Agreement) The algorthm satsfes Unform Agreement. Lemma 3.5 If at round r 1, a process p k receves messages (ready,w,faled ) and (ready j,w j,faled j ), such that ready and ready j are true, and {,j} J (r) k, then w = w j. Proof: [of Lemma 3.5] By nducton on r: The frst round where t can happen s r = 2. Processes p and p j can send a ready message at round 2 f and only f the frst round has been and j quescent, e f they have receved all the messages sent to them n the frst round. Consequently, W [] (1) = W j [j] (1), and w = w j. Now, we can prove that, f the lemma holds for all rounds before r, t also holds for round r. Suppose for convenence that p and p j become ready at rounds r and r j respectvely wth r rj. r < r j. Snce {, j} J (r) k, p j does not suspect p at round r j < r, so p j receved the ready message from p at round r j. Snce the lemma holds at round r j, all the ready messages accepted by p j carry the same value, so that W j [j] (rj) = w. r = r j. The rounds r and r j are and j quescent: Faled (r) = Faled (r 1) and Faled (r) j = Faled (r 1) j. Snce p and p j do not suspect each other (snce {, j} J (r) k ), both of them receved the messages from each other at round r, so that Faled (r 1) Faled (r) j and Faled (r 1) j Faled (r). Consequently, Faled (r) = Faled (rj) j : p and p j receved the same set of messages at round r, and thus, W [] (r) = Φ k (W (r) ) = Φ k (W (r) j ) = W j [j] (r), so that w = w j. Lemma 3.5 Proof: [of theorem 3.4] Let p be the frst process to decde, and p j another process that decdes later. p decdes v at round r, and p j decdes v j at round r j, wth r j. There are dfferent cases to consder: 5

7 1. r = t + 1 = r j (a) r = Faled (r) + 1, so Faled (r) = t,.e. p knows all ncorrect processes, and s correct.. r s quescent: Faled (r 1) = t. So, p s a correct, and ts r message contanng all the ncorrect processes s receved by all processes. Thus, all the processes that decde are correct, and they all know all ncorrect processes. If some of them were ready before round r, ther messages are receved and they carry the same value whch s decded by all processes. Otherwse, all processes receve the same messages, so they become ready wth the same value to decde.. r s not quescent: f one process s not ready at the end of round r 1 and stll decdes at round r, t must become ready at round r, and we fall back n the prevous case. So, suppose that all processes are ready at least at the end of r 1. All processes receve the ready messages sent at round r by all correct processes, together wth ther own ready message. From lemma 3.5, all these message carry the same value. (b) r = Faled (r) + 2. We can suppose that p s a correct process. Indeed, there s at least one correct process, and ether ths process satsfes the prevous case (and we can then use that case), or t satsfes ths case, snce all processes decde at round t + 1. Snce r = Faled (r) +2, p has at least seen two quescent rounds, thus, t was already ready at round r 1 = t. At round t + 1, all processes receve p message, snce p s correct, and thus must decde the same value (from lemma 3.5). 2. r < t + 1, so r = Faled (r) + 2. p has seen at least two quescent rounds before r, so t was ready at least at the end of round r 1. Let p c be a correct process. Faled c (r 1) Faled (r). Thus, Faled c (r 1) Faled (r) r 2 (r 1) 1. Consequently, p c has seen at least one quescent round before round r 1, and thus was ready at least at the end of round r 1. Durng round r, p receves both ts own ready message and the ready message from p c. From Lemma 3.5, r decdes the same value as p c. At round r j, p j receves also a ready message from p c, snce p c s correct. Consequently, p j decdes the same value as p (ether because t becomes ready at round r j, or because t receves ts own ready message and we can use the same Lemma). Theorem 3.4 We now address the valdty ssue. For that, we lmt the set to whch functons Φ k belong. More precsely, for each k {1,,n} we restrct ourselves to functons Φ k whch satsfy the followng (C k ) condton: (1) Φ k (W) = 0 { : W[] {0, }} k; (2) Φ k (W) {W[] : W[] } otherwse and W (,, ). Note that for every k {1,,n}, there exst functons whch satsfy (C k ). Moreover, part (2) n (C k ) s qute natural here snce Φ k s never appled to (,, ) durng the algorthm. We now prove that condton (C k ) for functon Φ k n Fgure 1 yelds an algorthm whch solves the k-tag problem. Theorem 3.6 (k-valdty) For any functon Φ k : (V { }) n V verfyng (C k ), the algorthm satsfes k-valdty. 6

8 Proof: Frst, t can be proved by a smple nducton that r, p, p j, W (r) [j] s ether or the ntal value of p j. Now, suppose that at least k processes start wth the ntal value 0. Let p 1,, p k denote such processes. For any round r 1 and process p, v [1..k], W (r) [p v ] {0, }; thus, Φ k (W (r) ) = 0, and 0 s the only possble decson value. On the other hand, consder a run of the algorthm n whch all the processes start wth 1, and wth at most k 1 falures. At the frst round, every process p receves at least n k + 1 messages wth value 1, and so r 1, Φ k (W (r) ) = 1. Theorem The General-Omsson Model states rounds N, ntally 0 faled N, ntally 0 W (V { }) n, ntally (,,v,, ) where v s p s ntal value halt, a Boolean, ntally false Faled Π, ntally decson V {unknown}, ntally unknown msgs f halt then send (decson,w) to all processes / Faled trans rounds := rounds + 1 let J be the set of ndces of processes, from whch a message has arrved let (decson j,w j ) be the message receved from p j, for each j J J := J\Faled Faled := Π\J f j J, decson j unknown, then decson := W j else f Faled > t then halt := true else j Π, W[j] := max{w k [j] k J } faled := Faled f j J, W j W, then faled := faled + 1 f rounds {faled + 2, Faled + 3,t + 1} then decson := Φ k (W) Fgure 2: An Early-Decdng Algorthm n the General-Omsson Model In ths secton, we study the lower-bounds for agreement n the general-omsson synchronous model: n ths model, an ncorrect process may perform send-omssons (fal to send some messages), receve-omssons (fal to receve messages correctly sent by other processes) or crash. 7

9 4.1 Lower-Bounds Frst of all, n the synchronous model wth general-omssons, we can present a frst bound: k-tag cannot be solved wthout a strct majorty of correct processes: Theorem 4.1 (Majorty) In the synchronous model wth general-omssons, k-tag cannot be solved f 2t n. Ths theorem s a smple case of the general result of Bazz and Neger [BN92], for coordnaton problems n synchronous systems wth general-omsson falures. Theorem 4.2 (Lower-Bounds) In the general-omsson model, for any k-tag algorthm, there s a run where one process needs mn(f +2,t+1) rounds to decde, f f s the number of falures n the run. Proof: Ths general-omsson model contans the send-omsson model. Thus, the bounds for sendomssons apply to the general-omssons. Theorem An Early-Decdng Algorthm Now, we ntroduce an algorthm that can be used to solve the k-tag problem n the generalomsson synchronous model. As n the prevous case, the algorthm s parameterzed by a functon Φ k, that can be nstancated to solve every k-tag problem. Ths algorthm acheves the lower-bounds of mn(f +2,t+1). The basc dea s to force detecton of ncorrect processes by breakng any communcaton between two processes when one of the two processes does not receve a message. A process can only suspect more than t processes f t s ncorrect, and thus, s forced to halt by the algorthm before decdng. Other ncorrect processes are not rejected, snce they receve enough nformaton to meet the agreement property from the majorty of correct processes. Contrary to our frst algorthm (for send-omssons), ths algorthm also uses a noton closed to quescent rounds, but here, t s used to delay the decson f the nformaton s not unformally dstrbuted among non-suspected processes. Theorem 4.3 (Unform Agreement) The algorthm satsfes Unform Agreement. Let p k be a process, we defne K (r) = {j W j [k] (r) halt (r) j } Lemma 4.4 Let p be a process and r > 0 a round. If / K (r ), halt (r ) and K (r ) Ø, then there exsts a lst (j r ) 0 r r 1, such that j r K (r) and all p jr are dstnct processes. Proof: [of Lemma 4.4] r [0..r ], K (r) Ø (a value can dsappear durng a round, and reappear after). Let r [1..r 1] be a round where K (r) K (r 1). Snce K (r+1) Ø, let p j K (r+1). 1. p j K (r) K (r 1) : Durng round r + 1, p and p j receved at least one message from the same process p j, snce they receved messages from a majorty of them durng ths round. Snce p j doesnot suspect p j at round r + 1, t receved ts vector W (r 1) j (contanng v k ) at round r, and forwarded ths value to p. Absurd snce p / K (r). 2. p j / K (r) : p j has receved the value v k durng round r+1 from a process p j K(r) K (r 1). Snce p j doesnot suspect p j at round r + 1, t receved a message from p j at round r that already contaned v k, thus p j K (r), contradcton. 8

10 Consequently, there s no round r [1..r 1] such that K (r) K (r 1). r [1..r 1], let j r be a process such that j r K (r) (non empty) and j r / K (r 1). Let j 0 K (0). Ths lst (j r ) satsfes the Lemma 4.4. Lemma 4.4 Proof: [of Theorem 4.3] Let r be the frst round where a process decdes. Let p be such a process, t decdes a vector W (r). Let p j be another process that decdes at round r j r on a dfferent vector W (rj) 1. r = r j : Let v k be the ntal value of p k such that K (r) and j / K (r). From Lemma 4.4, there s a lst (j r ) 0 r r 1 such that j r K (r) and all p jr are dstnct processes. Snce j / K (r), all these processes belong to Faled (r) j, so that Faled (r) j r. (a) r = t + 1: Faled (r) j t + 1, mpossble snce p j doesnot halt at round r. (b) r faled (r) j + 2: Faled (r) j faled (r) j j. + 2 Faled (r) j + 2, also mpossble. 2. r < r j : Let v k be the ntal value of p k such that K (r) and j / K (rj) (the opposte case can be shown mpossble by the same proof as above. (a) r = Faled (r) + 2: l J (r), W (r 1) l = W (r). At least one of these processes (they are a majorty) s receved by p j durng the round r, so p j already knows all the values n W (r) at the end of r. (b) r = Faled (r) + 3: If / K (r 2), from Lemma 4.4, there s a lst (j r ) 0 r r 3 such that j r K (r) and all p jr are dstnct processes. All these processes belong to Faled (r 2), so that Faled (r 2) r 2. But r = Faled (r) + 3, Faled (r 2) Faled (r) Faled (r) + 1 whch s mpossble. Thus, K (r 2), and at least on process trusted by p j receves v k from p durng round r 1, and forwards t to p j durng round r. Thus, p j cannot gnore v k. Theorem 4.3 Theorem 4.5 (Termnaton) The algorthm satsfes Termnaton. Proof: Clearly, at rount t + 1, all processes have fnally decded. Theorem 4.5 Theorem 4.6 (Optmalty) All correct processes decde n at most mn(f + 2, t + 1) rounds. Proof: If p s a correct process (t does not suspect a majorty of processes), t decdes at round f aled + 2, F aled + 3 or t + 1, or earler f t receves a message wth decde = true. We need to show that, f f < t 1, (1) there s a round r such that r {faled (r) + 2, Faled (r) + 3}. If r = Faled (r) + 3, we need also to show that (2) Faled (r) < f. (1) Suppose that there s no round r such that r = Faled (r) + 3. r mnmal such that r > Faled (r) +3. r 1 Faled (r 1) +3, snce r s mnmal. Snce Faled (r 1) Faled (r), we have Faled (r 1) + 3 Faled (r) + 3 < r Faled (r 1) + 4, e r = Faled (r 1) + 4, and thus r 1 = Faled (r 1) + 3. Absurd. Consequently, (1) s satsfed. 9

11 (2) r = Faled (r) + 3 (and r < r, r / {faled (r) + 2, Faled (r) + 3}). Snce r 1 Faled (r 1) + 3 and r 1 / {faled (r 1) + 2, Faled (r 1) + 3}, r 1 < Faled (r 1) + 3,.e. r < Faled (r 1) + 4. r = Faled (r) + 3 and r < Faled (r 1) + 4, wth Faled (r) Faled (r 1) mply that Faled (r 1) = Faled (r). Snce r 1 faled (r 1) + 2 and r 1 = Faled (r 1) + 2, there must exst p j such that j J (r 1) and W (r 2) j W (r 1), e p k, such that K (r 1) and j / K (r 2). From Lemma 4.4, there s a lst (j r ) 0 r r 3, of dstnct processes such that j r K (r). Snce j / K (r 2), all these processes belong to Faled (r 2) j Faled (r 1). (a) If p j s correct, these r 2 processes are correctly suspected,.e. f r 2 Faled (r) + 1. (b) If p j s not correct: j / Faled (r) and p s correct (t only suspects ncorrect processes), so that Faled (r) {j} s a subset of the set of ncorrect processes. Consequently, f Faled (r) + 1. In both cases, we have f Faled (r) + 1. Thus the decson s reached n less than f + 2 rounds. Theorem 4.6 Theorem 4.7 (k-valdty) For any functon Φ k : (V { }) n V verfyng (C k ), the algorthm satsfes k-valdty. Proof: The proof s exactly the same as for the k Valdty n the send-omsson model for the prevous algorthm (Theorem 3.6). Theorem Concluson We have presented two early-decdng algorthms for the k-tag class of agreement problems n the synchronous model wth message and crash falures. Both algorthms meet the known lower-bounds of mn(f + 2,t + 1) for early-decdng n ther partcular model: the frst algorthm can be used n the send-omsson model, whereas the second one s to be used n the general-omsson model. In ths latter model, agreement can only be reached f a majorty of processes are correct. These algorthms are partcularly appealng, snce they allow programs to reach agreement wth the best speed (n the number of synchronous rounds), whenever there are falures or not. References [BN92] [CL02] [CL03] R. Bazz and G. Neger. The possblty and the complexty of achevng faulttolerant coordnaton. In Prncples of dstrbuted computng (PODC 92), pages ACM Press, B. Charron-Bost and F. Le Fessant. Valdty condtons n agreement problemsand tme complexty. Techncal Report RR-4526, INRIA, Rocquencourt, August B. Charron-Bost and F. Le Fessant. Agreement problems reducton between models. In preparaton, January

12 [CS00] B. Charron-Bost and A. Schper. Unform consensus s harder than consensus. Techncal Report DSC/2000/028, EPFL, May [DFHR02] C. Delporte, H. Fauconner, J.-M. Helary, and M. Raynal. Early stoppng n global data computaton. Techncal Report 1436, IRISA, Jan [DRS86] [KR01] D. Dolev, R. Reschuk, and H. R. Strong. Early stoppng n Byzantne agreement. Techncal Report RJ5406, IBM Research Laboratory, December I. Kedar and S. Rajsbound. A smple proof of the unform consensus lower bound. July [Lyn96] N. A. Lynch. Dstrbuted Algorthms. Morgan Kaufmann, [Ros96a] [Ros96b] M.-C. Rosu. Early-stoppng termnatng relable broadcast protocol for generalomsson falures. In Prncples of dstrbuted computng (PODC 96), page 209. ACM Press, M.-C. Rosu. An optmal early-stoppng trb protocol for general-omsson falures. Techncal report, Cornell Unversty,

Problem Set 9 Solutions

Problem Set 9 Solutions Desgn and Analyss of Algorthms May 4, 2015 Massachusetts Insttute of Technology 6.046J/18.410J Profs. Erk Demane, Srn Devadas, and Nancy Lynch Problem Set 9 Solutons Problem Set 9 Solutons Ths problem

More information

Affine transformations and convexity

Affine transformations and convexity Affne transformatons and convexty The purpose of ths document s to prove some basc propertes of affne transformatons nvolvng convex sets. Here are a few onlne references for background nformaton: http://math.ucr.edu/

More information

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:

More information

Calculation of time complexity (3%)

Calculation of time complexity (3%) Problem 1. (30%) Calculaton of tme complexty (3%) Gven n ctes, usng exhaust search to see every result takes O(n!). Calculaton of tme needed to solve the problem (2%) 40 ctes:40! dfferent tours 40 add

More information

APPENDIX A Some Linear Algebra

APPENDIX A Some Linear Algebra APPENDIX A Some Lnear Algebra The collecton of m, n matrces A.1 Matrces a 1,1,..., a 1,n A = a m,1,..., a m,n wth real elements a,j s denoted by R m,n. If n = 1 then A s called a column vector. Smlarly,

More information

The Order Relation and Trace Inequalities for. Hermitian Operators

The Order Relation and Trace Inequalities for. Hermitian Operators Internatonal Mathematcal Forum, Vol 3, 08, no, 507-57 HIKARI Ltd, wwwm-hkarcom https://doorg/0988/mf088055 The Order Relaton and Trace Inequaltes for Hermtan Operators Y Huang School of Informaton Scence

More information

On the Multicriteria Integer Network Flow Problem

On the Multicriteria Integer Network Flow Problem BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 5, No 2 Sofa 2005 On the Multcrtera Integer Network Flow Problem Vassl Vasslev, Marana Nkolova, Maryana Vassleva Insttute of

More information

More metrics on cartesian products

More metrics on cartesian products More metrcs on cartesan products If (X, d ) are metrc spaces for 1 n, then n Secton II4 of the lecture notes we defned three metrcs on X whose underlyng topologes are the product topology The purpose of

More information

a b a In case b 0, a being divisible by b is the same as to say that

a b a In case b 0, a being divisible by b is the same as to say that Secton 6.2 Dvsblty among the ntegers An nteger a ε s dvsble by b ε f there s an nteger c ε such that a = bc. Note that s dvsble by any nteger b, snce = b. On the other hand, a s dvsble by only f a = :

More information

Lecture 4: November 17, Part 1 Single Buffer Management

Lecture 4: November 17, Part 1 Single Buffer Management Lecturer: Ad Rosén Algorthms for the anagement of Networs Fall 2003-2004 Lecture 4: November 7, 2003 Scrbe: Guy Grebla Part Sngle Buffer anagement In the prevous lecture we taled about the Combned Input

More information

find (x): given element x, return the canonical element of the set containing x;

find (x): given element x, return the canonical element of the set containing x; COS 43 Sprng, 009 Dsjont Set Unon Problem: Mantan a collecton of dsjont sets. Two operatons: fnd the set contanng a gven element; unte two sets nto one (destructvely). Approach: Canoncal element method:

More information

Introductory Cardinality Theory Alan Kaylor Cline

Introductory Cardinality Theory Alan Kaylor Cline Introductory Cardnalty Theory lan Kaylor Clne lthough by name the theory of set cardnalty may seem to be an offshoot of combnatorcs, the central nterest s actually nfnte sets. Combnatorcs deals wth fnte

More information

Difference Equations

Difference Equations Dfference Equatons c Jan Vrbk 1 Bascs Suppose a sequence of numbers, say a 0,a 1,a,a 3,... s defned by a certan general relatonshp between, say, three consecutve values of the sequence, e.g. a + +3a +1

More information

Perfect Competition and the Nash Bargaining Solution

Perfect Competition and the Nash Bargaining Solution Perfect Competton and the Nash Barganng Soluton Renhard John Department of Economcs Unversty of Bonn Adenauerallee 24-42 53113 Bonn, Germany emal: rohn@un-bonn.de May 2005 Abstract For a lnear exchange

More information

Edge Isoperimetric Inequalities

Edge Isoperimetric Inequalities November 7, 2005 Ross M. Rchardson Edge Isopermetrc Inequaltes 1 Four Questons Recall that n the last lecture we looked at the problem of sopermetrc nequaltes n the hypercube, Q n. Our noton of boundary

More information

Assortment Optimization under MNL

Assortment Optimization under MNL Assortment Optmzaton under MNL Haotan Song Aprl 30, 2017 1 Introducton The assortment optmzaton problem ams to fnd the revenue-maxmzng assortment of products to offer when the prces of products are fxed.

More information

CS : Algorithms and Uncertainty Lecture 17 Date: October 26, 2016

CS : Algorithms and Uncertainty Lecture 17 Date: October 26, 2016 CS 29-128: Algorthms and Uncertanty Lecture 17 Date: October 26, 2016 Instructor: Nkhl Bansal Scrbe: Mchael Denns 1 Introducton In ths lecture we wll be lookng nto the secretary problem, and an nterestng

More information

20. Mon, Oct. 13 What we have done so far corresponds roughly to Chapters 2 & 3 of Lee. Now we turn to Chapter 4. The first idea is connectedness.

20. Mon, Oct. 13 What we have done so far corresponds roughly to Chapters 2 & 3 of Lee. Now we turn to Chapter 4. The first idea is connectedness. 20. Mon, Oct. 13 What we have done so far corresponds roughly to Chapters 2 & 3 of Lee. Now we turn to Chapter 4. The frst dea s connectedness. Essentally, we want to say that a space cannot be decomposed

More information

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009 College of Computer & Informaton Scence Fall 2009 Northeastern Unversty 20 October 2009 CS7880: Algorthmc Power Tools Scrbe: Jan Wen and Laura Poplawsk Lecture Outlne: Prmal-dual schema Network Desgn:

More information

On the Interval Zoro Symmetric Single-step Procedure for Simultaneous Finding of Polynomial Zeros

On the Interval Zoro Symmetric Single-step Procedure for Simultaneous Finding of Polynomial Zeros Appled Mathematcal Scences, Vol. 5, 2011, no. 75, 3693-3706 On the Interval Zoro Symmetrc Sngle-step Procedure for Smultaneous Fndng of Polynomal Zeros S. F. M. Rusl, M. Mons, M. A. Hassan and W. J. Leong

More information

Module 9. Lecture 6. Duality in Assignment Problems

Module 9. Lecture 6. Duality in Assignment Problems Module 9 1 Lecture 6 Dualty n Assgnment Problems In ths lecture we attempt to answer few other mportant questons posed n earler lecture for (AP) and see how some of them can be explaned through the concept

More information

n α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0

n α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0 MODULE 2 Topcs: Lnear ndependence, bass and dmenson We have seen that f n a set of vectors one vector s a lnear combnaton of the remanng vectors n the set then the span of the set s unchanged f that vector

More information

Min Cut, Fast Cut, Polynomial Identities

Min Cut, Fast Cut, Polynomial Identities Randomzed Algorthms, Summer 016 Mn Cut, Fast Cut, Polynomal Identtes Instructor: Thomas Kesselhem and Kurt Mehlhorn 1 Mn Cuts n Graphs Lecture (5 pages) Throughout ths secton, G = (V, E) s a mult-graph.

More information

CS 331 DESIGN AND ANALYSIS OF ALGORITHMS DYNAMIC PROGRAMMING. Dr. Daisy Tang

CS 331 DESIGN AND ANALYSIS OF ALGORITHMS DYNAMIC PROGRAMMING. Dr. Daisy Tang CS DESIGN ND NLYSIS OF LGORITHMS DYNMIC PROGRMMING Dr. Dasy Tang Dynamc Programmng Idea: Problems can be dvded nto stages Soluton s a sequence o decsons and the decson at the current stage s based on the

More information

NP-Completeness : Proofs

NP-Completeness : Proofs NP-Completeness : Proofs Proof Methods A method to show a decson problem Π NP-complete s as follows. (1) Show Π NP. (2) Choose an NP-complete problem Π. (3) Show Π Π. A method to show an optmzaton problem

More information

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X Statstcs 1: Probablty Theory II 37 3 EPECTATION OF SEVERAL RANDOM VARIABLES As n Probablty Theory I, the nterest n most stuatons les not on the actual dstrbuton of a random vector, but rather on a number

More information

Common loop optimizations. Example to improve locality. Why Dependence Analysis. Data Dependence in Loops. Goal is to find best schedule:

Common loop optimizations. Example to improve locality. Why Dependence Analysis. Data Dependence in Loops. Goal is to find best schedule: 15-745 Lecture 6 Data Dependence n Loops Copyrght Seth Goldsten, 2008 Based on sldes from Allen&Kennedy Lecture 6 15-745 2005-8 1 Common loop optmzatons Hostng of loop-nvarant computatons pre-compute before

More information

Dynamic Programming. Preview. Dynamic Programming. Dynamic Programming. Dynamic Programming (Example: Fibonacci Sequence)

Dynamic Programming. Preview. Dynamic Programming. Dynamic Programming. Dynamic Programming (Example: Fibonacci Sequence) /24/27 Prevew Fbonacc Sequence Longest Common Subsequence Dynamc programmng s a method for solvng complex problems by breakng them down nto smpler sub-problems. It s applcable to problems exhbtng the propertes

More information

CHAPTER 17 Amortized Analysis

CHAPTER 17 Amortized Analysis CHAPTER 7 Amortzed Analyss In an amortzed analyss, the tme requred to perform a sequence of data structure operatons s averaged over all the operatons performed. It can be used to show that the average

More information

Iterative Approximate Byzantine Consensus in Arbitrary Directed Graphs

Iterative Approximate Byzantine Consensus in Arbitrary Directed Graphs Iteratve Approxmate Byzantne Consensus n Arbtrary Drected Graphs Ntn H. Vadya Unversty of Illnos Electrcal and Computer Engneerng Urbana, Illnos nhv@llnos.edu Lews Tseng Unversty of Illnos Computer Scence

More information

The Second Anti-Mathima on Game Theory

The Second Anti-Mathima on Game Theory The Second Ant-Mathma on Game Theory Ath. Kehagas December 1 2006 1 Introducton In ths note we wll examne the noton of game equlbrum for three types of games 1. 2-player 2-acton zero-sum games 2. 2-player

More information

Linear, affine, and convex sets and hulls In the sequel, unless otherwise specified, X will denote a real vector space.

Linear, affine, and convex sets and hulls In the sequel, unless otherwise specified, X will denote a real vector space. Lnear, affne, and convex sets and hulls In the sequel, unless otherwse specfed, X wll denote a real vector space. Lnes and segments. Gven two ponts x, y X, we defne xy = {x + t(y x) : t R} = {(1 t)x +

More information

Math 261 Exercise sheet 2

Math 261 Exercise sheet 2 Math 261 Exercse sheet 2 http://staff.aub.edu.lb/~nm116/teachng/2017/math261/ndex.html Verson: September 25, 2017 Answers are due for Monday 25 September, 11AM. The use of calculators s allowed. Exercse

More information

Foundations of Arithmetic

Foundations of Arithmetic Foundatons of Arthmetc Notaton We shall denote the sum and product of numbers n the usual notaton as a 2 + a 2 + a 3 + + a = a, a 1 a 2 a 3 a = a The notaton a b means a dvdes b,.e. ac = b where c s an

More information

The Minimum Universal Cost Flow in an Infeasible Flow Network

The Minimum Universal Cost Flow in an Infeasible Flow Network Journal of Scences, Islamc Republc of Iran 17(2): 175-180 (2006) Unversty of Tehran, ISSN 1016-1104 http://jscencesutacr The Mnmum Unversal Cost Flow n an Infeasble Flow Network H Saleh Fathabad * M Bagheran

More information

Learning Theory: Lecture Notes

Learning Theory: Lecture Notes Learnng Theory: Lecture Notes Lecturer: Kamalka Chaudhur Scrbe: Qush Wang October 27, 2012 1 The Agnostc PAC Model Recall that one of the constrants of the PAC model s that the data dstrbuton has to be

More information

Graph Reconstruction by Permutations

Graph Reconstruction by Permutations Graph Reconstructon by Permutatons Perre Ille and Wllam Kocay* Insttut de Mathémathques de Lumny CNRS UMR 6206 163 avenue de Lumny, Case 907 13288 Marselle Cedex 9, France e-mal: lle@ml.unv-mrs.fr Computer

More information

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg prnceton unv. F 17 cos 521: Advanced Algorthm Desgn Lecture 7: LP Dualty Lecturer: Matt Wenberg Scrbe: LP Dualty s an extremely useful tool for analyzng structural propertes of lnear programs. Whle there

More information

Lecture 10: May 6, 2013

Lecture 10: May 6, 2013 TTIC/CMSC 31150 Mathematcal Toolkt Sprng 013 Madhur Tulsan Lecture 10: May 6, 013 Scrbe: Wenje Luo In today s lecture, we manly talked about random walk on graphs and ntroduce the concept of graph expander,

More information

Week 5: Neural Networks

Week 5: Neural Networks Week 5: Neural Networks Instructor: Sergey Levne Neural Networks Summary In the prevous lecture, we saw how we can construct neural networks by extendng logstc regresson. Neural networks consst of multple

More information

Every planar graph is 4-colourable a proof without computer

Every planar graph is 4-colourable a proof without computer Peter Dörre Department of Informatcs and Natural Scences Fachhochschule Südwestfalen (Unversty of Appled Scences) Frauenstuhlweg 31, D-58644 Iserlohn, Germany Emal: doerre(at)fh-swf.de Mathematcs Subject

More information

The optimal delay of the second test is therefore approximately 210 hours earlier than =2.

The optimal delay of the second test is therefore approximately 210 hours earlier than =2. THE IEC 61508 FORMULAS 223 The optmal delay of the second test s therefore approxmately 210 hours earler than =2. 8.4 The IEC 61508 Formulas IEC 61508-6 provdes approxmaton formulas for the PF for smple

More information

Appendix B. The Finite Difference Scheme

Appendix B. The Finite Difference Scheme 140 APPENDIXES Appendx B. The Fnte Dfference Scheme In ths appendx we present numercal technques whch are used to approxmate solutons of system 3.1 3.3. A comprehensve treatment of theoretcal and mplementaton

More information

The Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction

The Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction ECONOMICS 5* -- NOTE (Summary) ECON 5* -- NOTE The Multple Classcal Lnear Regresson Model (CLRM): Specfcaton and Assumptons. Introducton CLRM stands for the Classcal Lnear Regresson Model. The CLRM s also

More information

Week 2. This week, we covered operations on sets and cardinality.

Week 2. This week, we covered operations on sets and cardinality. Week 2 Ths week, we covered operatons on sets and cardnalty. Defnton 0.1 (Correspondence). A correspondence between two sets A and B s a set S contaned n A B = {(a, b) a A, b B}. A correspondence from

More information

Supplement: Proofs and Technical Details for The Solution Path of the Generalized Lasso

Supplement: Proofs and Technical Details for The Solution Path of the Generalized Lasso Supplement: Proofs and Techncal Detals for The Soluton Path of the Generalzed Lasso Ryan J. Tbshran Jonathan Taylor In ths document we gve supplementary detals to the paper The Soluton Path of the Generalzed

More information

Composite Hypotheses testing

Composite Hypotheses testing Composte ypotheses testng In many hypothess testng problems there are many possble dstrbutons that can occur under each of the hypotheses. The output of the source s a set of parameters (ponts n a parameter

More information

Self-complementing permutations of k-uniform hypergraphs

Self-complementing permutations of k-uniform hypergraphs Dscrete Mathematcs Theoretcal Computer Scence DMTCS vol. 11:1, 2009, 117 124 Self-complementng permutatons of k-unform hypergraphs Artur Szymańsk A. Paweł Wojda Faculty of Appled Mathematcs, AGH Unversty

More information

Simultaneous Optimization of Berth Allocation, Quay Crane Assignment and Quay Crane Scheduling Problems in Container Terminals

Simultaneous Optimization of Berth Allocation, Quay Crane Assignment and Quay Crane Scheduling Problems in Container Terminals Smultaneous Optmzaton of Berth Allocaton, Quay Crane Assgnment and Quay Crane Schedulng Problems n Contaner Termnals Necat Aras, Yavuz Türkoğulları, Z. Caner Taşkın, Kuban Altınel Abstract In ths work,

More information

6. Stochastic processes (2)

6. Stochastic processes (2) Contents Markov processes Brth-death processes Lect6.ppt S-38.45 - Introducton to Teletraffc Theory Sprng 5 Markov process Consder a contnuous-tme and dscrete-state stochastc process X(t) wth state space

More information

6. Stochastic processes (2)

6. Stochastic processes (2) 6. Stochastc processes () Lect6.ppt S-38.45 - Introducton to Teletraffc Theory Sprng 5 6. Stochastc processes () Contents Markov processes Brth-death processes 6. Stochastc processes () Markov process

More information

NUMERICAL DIFFERENTIATION

NUMERICAL DIFFERENTIATION NUMERICAL DIFFERENTIATION 1 Introducton Dfferentaton s a method to compute the rate at whch a dependent output y changes wth respect to the change n the ndependent nput x. Ths rate of change s called the

More information

Perron Vectors of an Irreducible Nonnegative Interval Matrix

Perron Vectors of an Irreducible Nonnegative Interval Matrix Perron Vectors of an Irreducble Nonnegatve Interval Matrx Jr Rohn August 4 2005 Abstract As s well known an rreducble nonnegatve matrx possesses a unquely determned Perron vector. As the man result of

More information

Lecture Randomized Load Balancing strategies and their analysis. Probability concepts include, counting, the union bound, and Chernoff bounds.

Lecture Randomized Load Balancing strategies and their analysis. Probability concepts include, counting, the union bound, and Chernoff bounds. U.C. Berkeley CS273: Parallel and Dstrbuted Theory Lecture 1 Professor Satsh Rao August 26, 2010 Lecturer: Satsh Rao Last revsed September 2, 2010 Lecture 1 1 Course Outlne We wll cover a samplng of the

More information

Finding Dense Subgraphs in G(n, 1/2)

Finding Dense Subgraphs in G(n, 1/2) Fndng Dense Subgraphs n Gn, 1/ Atsh Das Sarma 1, Amt Deshpande, and Rav Kannan 1 Georga Insttute of Technology,atsh@cc.gatech.edu Mcrosoft Research-Bangalore,amtdesh,annan@mcrosoft.com Abstract. Fndng

More information

Example: (13320, 22140) =? Solution #1: The divisors of are 1, 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 27, 30, 36, 41,

Example: (13320, 22140) =? Solution #1: The divisors of are 1, 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 27, 30, 36, 41, The greatest common dvsor of two ntegers a and b (not both zero) s the largest nteger whch s a common factor of both a and b. We denote ths number by gcd(a, b), or smply (a, b) when there s no confuson

More information

VQ widely used in coding speech, image, and video

VQ widely used in coding speech, image, and video at Scalar quantzers are specal cases of vector quantzers (VQ): they are constraned to look at one sample at a tme (memoryless) VQ does not have such constrant better RD perfomance expected Source codng

More information

A combinatorial problem associated with nonograms

A combinatorial problem associated with nonograms A combnatoral problem assocated wth nonograms Jessca Benton Ron Snow Nolan Wallach March 21, 2005 1 Introducton. Ths work was motvated by a queston posed by the second named author to the frst named author

More information

MMA and GCMMA two methods for nonlinear optimization

MMA and GCMMA two methods for nonlinear optimization MMA and GCMMA two methods for nonlnear optmzaton Krster Svanberg Optmzaton and Systems Theory, KTH, Stockholm, Sweden. krlle@math.kth.se Ths note descrbes the algorthms used n the author s 2007 mplementatons

More information

Random Walks on Digraphs

Random Walks on Digraphs Random Walks on Dgraphs J. J. P. Veerman October 23, 27 Introducton Let V = {, n} be a vertex set and S a non-negatve row-stochastc matrx (.e. rows sum to ). V and S defne a dgraph G = G(V, S) and a drected

More information

Volume 18 Figure 1. Notation 1. Notation 2. Observation 1. Remark 1. Remark 2. Remark 3. Remark 4. Remark 5. Remark 6. Theorem A [2]. Theorem B [2].

Volume 18 Figure 1. Notation 1. Notation 2. Observation 1. Remark 1. Remark 2. Remark 3. Remark 4. Remark 5. Remark 6. Theorem A [2]. Theorem B [2]. Bulletn of Mathematcal Scences and Applcatons Submtted: 016-04-07 ISSN: 78-9634, Vol. 18, pp 1-10 Revsed: 016-09-08 do:10.1805/www.scpress.com/bmsa.18.1 Accepted: 016-10-13 017 ScPress Ltd., Swtzerland

More information

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS Avalable onlne at http://sck.org J. Math. Comput. Sc. 3 (3), No., 6-3 ISSN: 97-537 COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

More information

arxiv: v1 [math.co] 1 Mar 2014

arxiv: v1 [math.co] 1 Mar 2014 Unon-ntersectng set systems Gyula O.H. Katona and Dánel T. Nagy March 4, 014 arxv:1403.0088v1 [math.co] 1 Mar 014 Abstract Three ntersecton theorems are proved. Frst, we determne the sze of the largest

More information

Outline. Communication. Bellman Ford Algorithm. Bellman Ford Example. Bellman Ford Shortest Path [1]

Outline. Communication. Bellman Ford Algorithm. Bellman Ford Example. Bellman Ford Shortest Path [1] DYNAMIC SHORTEST PATH SEARCH AND SYNCHRONIZED TASK SWITCHING Jay Wagenpfel, Adran Trachte 2 Outlne Shortest Communcaton Path Searchng Bellmann Ford algorthm Algorthm for dynamc case Modfcatons to our algorthm

More information

Turing Machines (intro)

Turing Machines (intro) CHAPTER 3 The Church-Turng Thess Contents Turng Machnes defntons, examples, Turng-recognzable and Turng-decdable languages Varants of Turng Machne Multtape Turng machnes, non-determnstc Turng Machnes,

More information

Numerical Heat and Mass Transfer

Numerical Heat and Mass Transfer Master degree n Mechancal Engneerng Numercal Heat and Mass Transfer 06-Fnte-Dfference Method (One-dmensonal, steady state heat conducton) Fausto Arpno f.arpno@uncas.t Introducton Why we use models and

More information

Formulas for the Determinant

Formulas for the Determinant page 224 224 CHAPTER 3 Determnants e t te t e 2t 38 A = e t 2te t e 2t e t te t 2e 2t 39 If 123 A = 345, 456 compute the matrx product A adj(a) What can you conclude about det(a)? For Problems 40 43, use

More information

System in Weibull Distribution

System in Weibull Distribution Internatonal Matheatcal Foru 4 9 no. 9 94-95 Relablty Equvalence Factors of a Seres-Parallel Syste n Webull Dstrbuton M. A. El-Dacese Matheatcs Departent Faculty of Scence Tanta Unversty Tanta Egypt eldacese@yahoo.co

More information

Lecture Space-Bounded Derandomization

Lecture Space-Bounded Derandomization Notes on Complexty Theory Last updated: October, 2008 Jonathan Katz Lecture Space-Bounded Derandomzaton 1 Space-Bounded Derandomzaton We now dscuss derandomzaton of space-bounded algorthms. Here non-trval

More information

Compilers. Spring term. Alfonso Ortega: Enrique Alfonseca: Chapter 4: Syntactic analysis

Compilers. Spring term. Alfonso Ortega: Enrique Alfonseca: Chapter 4: Syntactic analysis Complers Sprng term Alfonso Ortega: alfonso.ortega@uam.es nrque Alfonseca: enrque.alfonseca@uam.es Chapter : Syntactc analyss. Introducton. Bottom-up Analyss Syntax Analyser Concepts It analyses the context-ndependent

More information

For now, let us focus on a specific model of neurons. These are simplified from reality but can achieve remarkable results.

For now, let us focus on a specific model of neurons. These are simplified from reality but can achieve remarkable results. Neural Networks : Dervaton compled by Alvn Wan from Professor Jtendra Malk s lecture Ths type of computaton s called deep learnng and s the most popular method for many problems, such as computer vson

More information

The L(2, 1)-Labeling on -Product of Graphs

The L(2, 1)-Labeling on -Product of Graphs Annals of Pure and Appled Mathematcs Vol 0, No, 05, 9-39 ISSN: 79-087X (P, 79-0888(onlne Publshed on 7 Aprl 05 wwwresearchmathscorg Annals of The L(, -Labelng on -Product of Graphs P Pradhan and Kamesh

More information

8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS

8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS SECTION 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS 493 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS All the vector spaces you have studed thus far n the text are real vector spaces because the scalars

More information

Basic Regular Expressions. Introduction. Introduction to Computability. Theory. Motivation. Lecture4: Regular Expressions

Basic Regular Expressions. Introduction. Introduction to Computability. Theory. Motivation. Lecture4: Regular Expressions Introducton to Computablty Theory Lecture: egular Expressons Prof Amos Israel Motvaton If one wants to descrbe a regular language, La, she can use the a DFA, Dor an NFA N, such L ( D = La that that Ths

More information

= z 20 z n. (k 20) + 4 z k = 4

= z 20 z n. (k 20) + 4 z k = 4 Problem Set #7 solutons 7.2.. (a Fnd the coeffcent of z k n (z + z 5 + z 6 + z 7 + 5, k 20. We use the known seres expanson ( n+l ( z l l z n below: (z + z 5 + z 6 + z 7 + 5 (z 5 ( + z + z 2 + z + 5 5

More information

Lecture 12: Discrete Laplacian

Lecture 12: Discrete Laplacian Lecture 12: Dscrete Laplacan Scrbe: Tanye Lu Our goal s to come up wth a dscrete verson of Laplacan operator for trangulated surfaces, so that we can use t n practce to solve related problems We are mostly

More information

THE CHINESE REMAINDER THEOREM. We should thank the Chinese for their wonderful remainder theorem. Glenn Stevens

THE CHINESE REMAINDER THEOREM. We should thank the Chinese for their wonderful remainder theorem. Glenn Stevens THE CHINESE REMAINDER THEOREM KEITH CONRAD We should thank the Chnese for ther wonderful remander theorem. Glenn Stevens 1. Introducton The Chnese remander theorem says we can unquely solve any par of

More information

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification E395 - Pattern Recognton Solutons to Introducton to Pattern Recognton, Chapter : Bayesan pattern classfcaton Preface Ths document s a soluton manual for selected exercses from Introducton to Pattern Recognton

More information

ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EQUATION

ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EQUATION Advanced Mathematcal Models & Applcatons Vol.3, No.3, 2018, pp.215-222 ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EUATION

More information

A New Refinement of Jacobi Method for Solution of Linear System Equations AX=b

A New Refinement of Jacobi Method for Solution of Linear System Equations AX=b Int J Contemp Math Scences, Vol 3, 28, no 17, 819-827 A New Refnement of Jacob Method for Soluton of Lnear System Equatons AX=b F Naem Dafchah Department of Mathematcs, Faculty of Scences Unversty of Gulan,

More information

Voting Games with Positive Weights and. Dummy Players: Facts and Theory

Voting Games with Positive Weights and. Dummy Players: Facts and Theory Appled Mathematcal Scences, Vol 10, 2016, no 53, 2637-2646 HIKARI Ltd, wwwm-hkarcom http://dxdoorg/1012988/ams201667209 Votng Games wth Postve Weghts and Dummy Players: Facts and Theory Zdravko Dmtrov

More information

FACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP

FACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP C O L L O Q U I U M M A T H E M A T I C U M VOL. 80 1999 NO. 1 FACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP BY FLORIAN K A I N R A T H (GRAZ) Abstract. Let H be a Krull monod wth nfnte class

More information

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons

More information

THERE ARE INFINITELY MANY FIBONACCI COMPOSITES WITH PRIME SUBSCRIPTS

THERE ARE INFINITELY MANY FIBONACCI COMPOSITES WITH PRIME SUBSCRIPTS Research and Communcatons n Mathematcs and Mathematcal Scences Vol 10, Issue 2, 2018, Pages 123-140 ISSN 2319-6939 Publshed Onlne on November 19, 2018 2018 Jyot Academc Press http://jyotacademcpressorg

More information

Complete subgraphs in multipartite graphs

Complete subgraphs in multipartite graphs Complete subgraphs n multpartte graphs FLORIAN PFENDER Unverstät Rostock, Insttut für Mathematk D-18057 Rostock, Germany Floran.Pfender@un-rostock.de Abstract Turán s Theorem states that every graph G

More information

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix Lectures - Week 4 Matrx norms, Condtonng, Vector Spaces, Lnear Independence, Spannng sets and Bass, Null space and Range of a Matrx Matrx Norms Now we turn to assocatng a number to each matrx. We could

More information

12 MATH 101A: ALGEBRA I, PART C: MULTILINEAR ALGEBRA. 4. Tensor product

12 MATH 101A: ALGEBRA I, PART C: MULTILINEAR ALGEBRA. 4. Tensor product 12 MATH 101A: ALGEBRA I, PART C: MULTILINEAR ALGEBRA Here s an outlne of what I dd: (1) categorcal defnton (2) constructon (3) lst of basc propertes (4) dstrbutve property (5) rght exactness (6) localzaton

More information

EEL 6266 Power System Operation and Control. Chapter 3 Economic Dispatch Using Dynamic Programming

EEL 6266 Power System Operation and Control. Chapter 3 Economic Dispatch Using Dynamic Programming EEL 6266 Power System Operaton and Control Chapter 3 Economc Dspatch Usng Dynamc Programmng Pecewse Lnear Cost Functons Common practce many utltes prefer to represent ther generator cost functons as sngle-

More information

Grover s Algorithm + Quantum Zeno Effect + Vaidman

Grover s Algorithm + Quantum Zeno Effect + Vaidman Grover s Algorthm + Quantum Zeno Effect + Vadman CS 294-2 Bomb 10/12/04 Fall 2004 Lecture 11 Grover s algorthm Recall that Grover s algorthm for searchng over a space of sze wors as follows: consder the

More information

Problem Do any of the following determine homomorphisms from GL n (C) to GL n (C)?

Problem Do any of the following determine homomorphisms from GL n (C) to GL n (C)? Homework 8 solutons. Problem 16.1. Whch of the followng defne homomomorphsms from C\{0} to C\{0}? Answer. a) f 1 : z z Yes, f 1 s a homomorphsm. We have that z s the complex conjugate of z. If z 1,z 2

More information

General theory of fuzzy connectedness segmentations: reconciliation of two tracks of FC theory

General theory of fuzzy connectedness segmentations: reconciliation of two tracks of FC theory General theory of fuzzy connectedness segmentatons: reconclaton of two tracks of FC theory Krzysztof Chrs Ceselsk Department of Mathematcs, West Vrgna Unversty and MIPG, Department of Radology, Unversty

More information

Neryškioji dichotominių testo klausimų ir socialinių rodiklių diferencijavimo savybių klasifikacija

Neryškioji dichotominių testo klausimų ir socialinių rodiklių diferencijavimo savybių klasifikacija Neryškoj dchotomnų testo klausmų r socalnų rodklų dferencjavmo savybų klasfkacja Aleksandras KRYLOVAS, Natalja KOSAREVA, Julja KARALIŪNAITĖ Technologcal and Economc Development of Economy Receved 9 May

More information

Amusing Properties of Odd Numbers Derived From Valuated Binary Tree

Amusing Properties of Odd Numbers Derived From Valuated Binary Tree IOSR Journal of Mathematcs (IOSR-JM) e-iss: 78-578, p-iss: 19-765X. Volume 1, Issue 6 Ver. V (ov. - Dec.016), PP 5-57 www.osrjournals.org Amusng Propertes of Odd umbers Derved From Valuated Bnary Tree

More information

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4)

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4) I. Classcal Assumptons Econ7 Appled Econometrcs Topc 3: Classcal Model (Studenmund, Chapter 4) We have defned OLS and studed some algebrac propertes of OLS. In ths topc we wll study statstcal propertes

More information

Lecture 21: Numerical methods for pricing American type derivatives

Lecture 21: Numerical methods for pricing American type derivatives Lecture 21: Numercal methods for prcng Amercan type dervatves Xaoguang Wang STAT 598W Aprl 10th, 2014 (STAT 598W) Lecture 21 1 / 26 Outlne 1 Fnte Dfference Method Explct Method Penalty Method (STAT 598W)

More information

Valuated Binary Tree: A New Approach in Study of Integers

Valuated Binary Tree: A New Approach in Study of Integers Internatonal Journal of Scentfc Innovatve Mathematcal Research (IJSIMR) Volume 4, Issue 3, March 6, PP 63-67 ISS 347-37X (Prnt) & ISS 347-34 (Onlne) wwwarcournalsorg Valuated Bnary Tree: A ew Approach

More information

Lecture 20: Lift and Project, SDP Duality. Today we will study the Lift and Project method. Then we will prove the SDP duality theorem.

Lecture 20: Lift and Project, SDP Duality. Today we will study the Lift and Project method. Then we will prove the SDP duality theorem. prnceton u. sp 02 cos 598B: algorthms and complexty Lecture 20: Lft and Project, SDP Dualty Lecturer: Sanjeev Arora Scrbe:Yury Makarychev Today we wll study the Lft and Project method. Then we wll prove

More information

Homotopy Type Theory Lecture Notes

Homotopy Type Theory Lecture Notes 15-819 Homotopy Type Theory Lecture Notes Evan Cavallo and Stefan Muller November 18 and 20, 2013 1 Reconsder Nat n smple types s a warmup to dscussng nductve types, we frst revew several equvalent presentatons

More information

Ali Omer Alattass Department of Mathematics, Faculty of Science, Hadramout University of science and Technology, P. O. Box 50663, Mukalla, Yemen

Ali Omer Alattass Department of Mathematics, Faculty of Science, Hadramout University of science and Technology, P. O. Box 50663, Mukalla, Yemen Journal of athematcs and Statstcs 7 (): 4448, 0 ISSN 5493644 00 Scence Publcatons odules n σ[] wth Chan Condtons on Small Submodules Al Omer Alattass Department of athematcs, Faculty of Scence, Hadramout

More information

A Hybrid Variational Iteration Method for Blasius Equation

A Hybrid Variational Iteration Method for Blasius Equation Avalable at http://pvamu.edu/aam Appl. Appl. Math. ISSN: 1932-9466 Vol. 10, Issue 1 (June 2015), pp. 223-229 Applcatons and Appled Mathematcs: An Internatonal Journal (AAM) A Hybrd Varatonal Iteraton Method

More information