A POLYNOMIAL-SPACE EXACT ALGORITHM FOR TSP IN DEGREE-5 GRAPHS

Size: px
Start display at page:

Download "A POLYNOMIAL-SPACE EXACT ALGORITHM FOR TSP IN DEGREE-5 GRAPHS"

Transcription

1 A POLYNOMIAL-SPACE EXACT ALGORITHM FOR TSP IN DEGREE-5 GRAPHS Norhazwani Md Yunos 1,2, Alksandar Shurbski 1, Hiroshi Nagamochi 1 1 Dpartmnt of Applid Mathmatics and Physics, Graduat School of Informatics, Kyoto Unirsity, , Kyoto, Japan 2 Unirsiti Tknikal Malaysia Mlaka, Durian Tunggal, Mlaka, Malaysia {wani, shurbski, nag}@amp.i.kyoto-u.ac.jp Kywords: Traling Salsman Problm, Exact Exponntial Algorithm, Branch-and-rduc, Masurand-conqur. Abstract Th Traling Salsman Problm (TSP) is on of th most wll-known NP-hard optimization problms. Following a rcnt trnd of rsarch which focuss on dloping algorithms for spcial typs of TSP instancs, namly graphs of limitd dgr, and thus alliating a part of th tim and spac complxity, w prsnt a polynomial-spac branching algorithm for th TSP in graphs with dgr at mos, and show that it has a running tim of O ( n ). To th bst of our knowldg, this is th first xact algorithm spcializd to graphs of such high dgr. Whil th bas of th xponnt in th running tim bound is gratr than two, our algorithm uss spac mrly polynomial in an input instanc siz, and thus by far outprforms Gurich and Shlah s O (4 n n log n ) polynomial-spac xact algorithm for th gnral TSP (Siam Journal of Computation, Vol. 16, No. 3, pp , 1987). In th analysis of th running tim, w us th masur-andconqur mthod, and w dlop a st of branching ruls which fostr th analysis of th running tim. 1 Introduction Th Traling Salsman Problm (TSP) is on of th most xtnsily studid problms in optimization. It has bn formulatd as a mathmatical problm in th 1930s. Many algorithmic mthods ha bn instigatd to bat th challng of finding th fastst algorithm in trms of running tim. On th othr hand, it has pron n mor challnging to dis fast algorithms that would us a managabl amount of computation spac, boundd by a polynomial in an input instanc s siz. W will riw prious algorithmic attmpts, making a distinction btwn thos which rquir spac xponntial in th siz of a problm instanc, and thos rquiring spac mrly polynomial in th input siz. W us th O notation, which supprsss polynomial factors. Th first non-triial algorithm for th TSP in an n- rtx graph is th O (2 n )-tim dynamic programming algorithm discord indpndntly by Bllman [1], and Hld and Karp [9] in th arly 1960s. This dynamic programming algorithm howr, rquirs also an xponntial amount of spac. Er sinc, this running tim has only bn improd for spcial typs of graphs. Primarily, instigation fforts ha bn focusd on graphs in which rtics ha a limitd dgr. Hncforth, lt dgr-i graph stand for a graph in which rtics ha maximum dgr at most i. A rcnt impromnt of th tim bound to O ( n ) for dgr-3 graphs has bn prsntd by Bodlandr t al. [2]. Thy ha usd a gnral approach for spding up straightforward dynamic programming algorithms. For TSP in dgr-4 graphs, Gbaur [7] has shown a tim bound of O (1.733 n ), by using a dynamic programming approach. In th in of polynomial spac algorithms, Gurich and Shlah [8] ha shown that th TSP in a gnral n-rtx graph is solabl in tim O ( 4 n n log n). Eppstin [4] has startd th xploration into polynomial spac TSP algorithms spcializd for graphs of boundd dgr by dsigning an algorithm for dgr-3 graphs that runs in O (1.260 n )-tim. H introducd a branch-and-sarch mthod by considring a gnralization of th TSP calld th forcd TSP. Iwama and Nakashima [10] ha claimd an impromnt of Eppstin s tim bound to O (1.251 n )-tim for TSP in dgr-3 graphs. Latr, Liskiwicz and Schustr [11] ha uncord som orsights mad in Iwama and Nakashima s analysis, and prod that thir algorithm actually runs in O (1.257 n )-tim. Liskiwicz and Schustr thn mad som minor modifications of Eppstin s algorithm and showd that this modifid algorithm runs in O ( n )-tim, a slight impromnt or Iwama and Nakashima s algorithm. Xiao and Nagamochi [14] ha rcntly prsntd an O ( n )-tim algorithm for TSP in dgr-3 graphs, and this improd prious tim bounds for polynomial-spac algorithms. Thy usd th basic stps of Eppstin s branch-and-sarch algorithm, and introducd a branching rul basd on a cut-circuit structur. In th procss of improing th tim bound, thy usd simpl analysis of masur and conqur, and ffctily analyzd thir algorithm by introducing an amortization schm or th cut circuit structur, stting wights to both rtics and connctd compo ISORA IET 45 Luoyang, China, Augus1 24, 2015

2 nnts of inducd graphs. For TSP in dgr-4 graphs, Eppstin [4] dsignd an algorithm that runs in O (1.890 n )-tim, basd on a branch-and-sarch mthod. Latr, Xiao and Nagamochi [13] showd an improd alu for th uppr bound of th running tim and showd that thir algorithm runs in O (1.692 n )-tim. Currntly, this is th fastst algorithm for TSP in dgr-4 graphs. Basically, th ida bhind thir algorithm is to apply rduction ruls until no furthr rduction is possibl, and thn branch on an dg by ithr including it to a solution or xcluding it from a solution. This is similar to most prious branch-and-sarch algorithms for th TSP. To ffctily analyz thir algorithm, Xiao and Nagamochi usd th masur and conqur mthod by stting a wight to ach rtx in a graph. From ach branching opration, thy drid a branching ctor using th assignd wight and aluat how much wight can b dcrasd in ach of th two instancs obtaind by branching on a slctd dg. In this way, thy wr abl to analyz by how much th total wight would dcras in ach branch. Moror, thy indicatd that th masur will dcras mor if w slct a good dg to branch on, and ga a st of simpl ruls, basd on a graph s topological proprtis, for choosing such an dg. Howr, th analysis of th running tim itslf is not as straightforward [13]. To th bst of our knowldg, thr xist no rports in th litratur of xact algorithms spcializd to th TSP in dgr-5 graphs. Thrfor, this papr prsnts th first algorithm for th TSP in dgr-5 graphs, and prsnts an uppr bound on th running tim of O ( n ). In this xploration, w us a dtrministic branch-and-sarch algorithm for TSP in dgr-5 graphs. Basically, our algorithm mploys similar tchniqus to most prious branching algorithms for th TSP. Whn thr ar no rtics of dgr 5 in an input graph, w call an xisting algorithm for TSP in dgr-4 graphs, and sol th rmaining instanc. In th analysis, w us th masur and conqur mthod as a tool to gt an uppr bound of th running tim. Th rmaindr of this papr is organizd as follows; Sction 2 oriws th basic notation usd in this papr and prsnts an introduction to th branching algorithm and masur and conqur mthod. Sction 3 dscribs our polynomial-spac branching algorithm. W stat our main rsult in sction 4, whr w procd with th analysis of th proposd algorithm. Finally, Sction 5 concluds th papr. 2 Mthods 2.1 Prliminaris For a graph G, lt V (G) dnot th st of rtics in G, and lt E(G) dnot th st of dgs in G. A pair of rtics and u ar calld nighbors if and u ar adjacnt by an dg u. W dnot th st of all nighbors of a rtx by N(), and dnot by d() th cardinality N() of N(), also calld th dgr of. For a subst of rtics W V (G), lt N(; W ) = N() W. For a subst of dgs E E(G), lt N E () = N() {u u E }, and lt d E () = N E (). Analogously, lt N E (; W ) = N E () W, and d E (, W ) = N E (, W ). Also, for a subst E of E(G), w dnot by G E th graph (V, E \ E ) obtaind from G by rmoing th dgs in E. W considr a gnralization of th TSP, namd th forcd Traling Salsman Problm. W dfin an instanc I = (G, F ) that consists of a simpl, dg wightd, undirctd graph G, and a subst F of dgs in G, calld forcd. A rtx is calld forcd if xactly on of its incidnt dgs is forcd. Similarly, it is calld unforcd if no forcd dg is incidnt to it. A Hamiltonian cycl in G is calld a tour if it passs through all th forcd dgs in F. Undr ths circumstancs, th forcd TSP rqusts to find a minimum cost tour of an instanc (G, F ). In this papr, w assum that th maximum dgr of a rtx in G is at mos. W dnot a forcd (rsp., unforcd) rtx of dgr i by fi (rsp., ui). W ar intrstd in six typs of rtics in an instanc of (G, F ), namly, u5, f5, u4, f4, u3 and f3-rtics. As shall b sn in Subsction 2.4.1, forcd and unforcd rtics of dgr 2 and 1 ar tratd as spcial cass. Lt V fi (rsp., V ui ), i = 3, 4, 5 dnot th st of firtics (rsp., ui-rtics) in (G, F ). 2.2 Essntials on Branching Algorithms W hr riw how to dri an uppr bound on th numbr of instancs that can b gnratd from an initial instanc by a branching algorithm. W can rprsnt th solution spac in our branching algorithm as a sarch tr. This is a ry usful way to illustrat th xcution of th branching ruls, and to aid th tim analysis of th branching algorithm. Th sarch tr is obtaind by assigning th input instanc of a problm as a root nod, and rcursily assigning a child to a nod for ach smallr instanc obtaind by applying th branching ruls. For a singl nod of th sarch tr, th algorithm taks tim polynomial in th siz of th nod instanc, which in turn, is smallr than or qual to th original instanc siz. Thus, w can conclud that th running tim of th branching algorithm is qual to th numbr of nods of th sarch tr tims a polynomial of th original input instanc siz. Lt I b a gin instanc with siz µ, and lt I and I b instancs obtaind from I by a branching opration. W us T (µ) to dnot th maximum numbr of nods in th sarch tr of an input of siz µ whn w xcut our branching algorithm. Lt a and b b th amounts of dcras in siz of instancs I and I, rspctily; ths alus dirctly dtrmin th prformanc of th algorithm. Thn, w call (a, b) th 2015 ISORA IET 46 Luoyang, China, Augus1 24, 2015

3 branching ctor of th branching ruls, and this implis th linar rcurrnc: T (µ) T (µ a) + T (µ b). (1) To aluat th prformanc of this branching ctor, w can us any standard mthod for linar rcurrnc rlations. In fact, it is known that T (µ) is of th form O (τ µ ), whr τ is th uniqu positi ral root of th function f(x) = 1 ( x a + x b) [6]. Th alu τ is calld th branching factor (of a gin branching ctor), and th running tim of th algorithm dcrass with th alu of this branching factor. 2.3 Th Masur-and-Conqur Mthod To ffctily analyz our sarch tr algorithm, w us th masur and conqur mthod. A complt dscription of this mthod is byond th scop of this papr, and th intrstd radr might rfr to th book of Fomin and Kratsch [6]. Th basic ida bhind th masur and conqur mthod is to assign a masur to an instanc, as opposd to using simply its siz whn analyzing th branching ctors of th branching oprations. A good choic for a masur might lad to a significantly improd analysis on th uppr bound of th running tim of a branching algorithm. For xampl, Fomin t al. [5] ha prsntd simpl polynomial-spac algorithms for th Maximum Indpndnt St and th Minimum Dominating St Problm, and obtaind an imprssi rfinmnt of th tim analysis by using th masur and conqur mthod. This shows that a good choic of masur is ry important to th tim bounds achiabl. For a gin problm instanc I of siz µ, lt W (I) b th masur of I. Whn considring a branch and rduc algorithm for th concrnd problm, intuitily w sk for a masur which satisfis th following proprtis (i) W (I) = 0 if and only if I can b sold in polynomial tim; (ii) If I is a sub-instanc of I obtaind through a rduction or a branching opration, thn W (I ) W (I). W call a masur W satisfying conditions (i) and (ii) abo a propr masur. 2.4 A Polynomial-Spac Branching Algorithm W assum that th maximum dgr of a rtx in a gin graph G is at mos. Basically, our algorithm contains two major stps. In th first stp, th algorithm applis rduction ruls until no furthr rduction is possibl. In th scond stp, th algorithm applis branching ruls in a rducd instanc to sarch for a solution. Ths two stps ar rpatd itratily. As a rsult of th rduction and branching oprations, th dgr of som rtics will dcras, whil th dgr of othr rtics will rmain unchangd. A forcd dg will nr disappar, nithr by th rduction nor branching oprations, but an unforcd dg may b rasd by ithr of th rduction or branching opration. Throughout th procss of th rduction and branching oprations, th masur of an instanc will nr incras. Dtails about th rduction and branching procdurs will b discussd in th following sub-sctions Rduction Ruls Rduction is a procss of transforming an instanc to a smallr instanc. It taks polynomial-tim to obtain a solution of an original instanc from a solution of a smallr instanc that has bn obtaind by a rduction procdur from th original instanc. Not all forcd TSP instancs ha a tour. If an instanc has no tour, w calld it infasibl. Lmma 1 gis two sufficint conditions for an instanc to b infasibl. Lmma 1 If on of th following conditions holds, thn th instanc (G, F ) is infasibl. (i) d() 1 for som rtx V (G). (ii) d F () 3 for som rtx V (G). In this papr, thr ar two rduction ruls applid in ach of th branching opration. Ths rduction ruls prsr th minimum cost tour of an instanc, as statd in Lmma 2. Lmma 2 Each of th following rductions prsrs th fasibility and a minimum cost tour of an instanc (G, F ). (i) If d() = 2 for a rtx, thn add to F any unforcd dg incidnt to rtx ; and (ii) If d() > 2 and d F () = 2 for a rtx, thn rmo from G any unforcd dg incidnt to rtx. Proof. Statmnts (i) and (ii) immdiatly follow from th dfinition of tours. From Lmma 1 and Lmma 2, w form our rduction algorithm as dscribd in Figur 1. An instanc (G, F ) which dos not satisfy any of th conditions in Lmma 1 and Lmma 2 is calld rducd Branching Ruls Our algorithm itratily branchs on an unforcd dg in a rducd instanc I = (G, F ) by ithr including into F, forc(), or xcluding it from G, dlt(). By applying a branching opration, th algorithm gnrats two nw instancs, calld branchs, by adding an unforcd dg to F, or by rmoing it from G ISORA IET 47 Luoyang, China, Augus1 24, 2015

4 Input: An instanc (G, F ) such that th maximum dgr of G is at mos. Output: A mssag for th infasibility of (G, F ); or a rducd instanc (G, F ) of (G, F ). Initializ (G, F ) := (G, F ); whil (G, F ) is not a rducd instanc do If thr is a rtx in (G, F ) such that d() 1 or d F () 3 thn Rturn mssag Infasibl Elsif thr is a rtx in (G, F ) such tha = d() > d F () thn Lt E b th st of unforcd dgs incidnt to all such rtics; St F := F E Elsif thr is a rtx in (G, F ) such that d() > d F () = 2 thn Lt E b th st of unforcd dgs incidnt to all such rtics; St G := G E End whil; Rturn (G, F ). Figur 1: Algorithm Rd(G, F ) To dscrib our branching algorithm, lt (G, F ) b a rducd instanc such that th maximum dgr of G is at mos. In (G, F ), an unforcd dg = t incidnt to a rtx of dgr 5 is calld optimal, if it satisfis a condition (c-i) blow with minimum indx i, or all unforcd dgs t in (G, F ): (c-1) V f5 and t N U (; V f3 ) such that N U () N U (t) = ; (c-2) V f5 and t N U (; V f3 ) such that N U () N U (t) ; (c-3) V f5 and t N U (; V u3 ); (c-4) V f5 and t N U (; V f4 ) such that N U () N U (t) = ; (c-5) V f5 and t N U (; V f4 ) such that N U () N U (t) ; (I) N U () N U (t) = 1; and (II) N U () N U (t) = 2; (c-6) V f5 and t N U (; V u4 ); (c-7) V f5 and t N U (; V f5 ) such that N U () N U (t) = ; (c-8) V f5 and t N U (; V f5 ) such that N U () N U (t) ; (I) N U () N U (t) = 1; (II) N U () N U (t) = 2; and (III) N U () N U (t) = 3; (c-9) V f5 and t N U (; V u5 ); (c-10) V u5 and t N U (; V f3 ); (c-11) V u5 and t N U (; V f4 ); (c-12) V u5 and t N U (; V u3 ); (c-13) V u5 and t N U (; V u4 ); and (c-14) V u5 and t N U (; V u5 ). W rfr to th abo conditions for choosing an optimal dg to branch on, c-1 to c-14, as th branching ruls. Th collcti st of branching ruls ar illustratd in Figur 2. For conninc in th analysis of th algorithm, cas (c-5) and cas (c-8) ha bn subdiidd into subcass according to th cardinality of th nighborhood intrsction. Intrsctions of lowr cardinality tak prcdnc or highr ons. Gin a rducd instanc I = (G, F ), our algorithm first chcks whthr thr xists a rtx of dgr 5, and if it dos, chooss an optimal dg according to th branching ruls. If thr xists no optimal dg according to th branching ruls, thn th rducd instanc has no mor rtics of dgr 5, and th maximum dgr of th rducd instanc at this point is at mos. Thn, w can call a polynomial spac xact algorithm for th TSP that is spcializd for dgr-4 graphs,.g., th algorithm spcializd for dgr-4 graphs by Xiao and Nagamochi [13]. Our branching algorithm is dscribd in Figur Wight Stting In ordr to obtain a masur which will imply th sam running tim bound as a function of th siz of a TSP instanc, w rquir that th wight of ach rtx is not gratr than 1. In what follows, w xamin som ncssary constraints which th rtx wights should satisfy in ordr to obtain a propr masur. For i = {3, 4, 5}, w dnot w i to b th wight of a ui-rtx, and w i to b th wight of an fi-rtx. Th conditions for a propr masur rquir that th masur of an instanc obtaind through a branching or a rduction opration will not b gratr than th masur of th original instanc. Thus, rtx wights should satisfy th following rlations w 5 1, (2) w 5 w 5, (3) w 4 w 4, (4) w 3 w 3, (5) w 3 w 4 w 5, and (6) w 3 w 4 w 5. (7) Th rtx wight for rtics of dgr lss than 3 is st to b 0. W procd to show in th algorithms gin in Figurs 1 and 3, stting rtx wights which satisfy th conditions of Eqs. (3) to (7) is sufficint to obtain a propr masur. Lmma 3 If th wights of rtics ar chosn as in Eqs. (3) to (7), thn th masur W (I) nr incrass as a rsult of th rduction or th branching oprations of Figur 1 and Figur 3. Proof. Lt I = (G, F ) b a gin instanc of th forcd TSP. Du to our dfinition of th masur W (I) of 2015 ISORA IET 48 Luoyang, China, Augus1 24, 2015

5 c-1 c-2 t 6 t 6 c-3 c-4 c-5(i) c-5(ii) t 7 t6 t6 c-6 c-7 t 7 t2 t3 t 6 c-8(i) c-8(ii) c-8(iii) t 8 t6 t 7 c-9 t 6 c-10 t 6 t 7 c-11 t 6 t 7 c-12 t 6 t7 t 8 c-13 t t 6 9 t7 t 8 c-14 : unforcd dgs : forcd dgs Figur 2: Illustration of th Branching Ruls Eq.(18), it suffics to show that non of th indiidual rtx wights will incras as a rsult of a rduction or a branching opration in th algorithms of Figurs 1 and 3. Th branching ruls stat that for an unforcd dg in E(G)\F, two subinstancs ar gnratd by ithr 2015 ISORA IET 49 Luoyang, China, Augus1 24, 2015

6 Input: An instanc (G, F ) such that th maximum dgr of G is at mos. Output: A mssag for th infasibility of (G, F ); or th minimum cost of a tour of (G, F ). Run Rd(G, F ); If Rd(G, F ) rturns mssag Infasibl thn Rturn mssag Infasibl Els Lt (G, F ) := Rd(G, F ); If V u5 V f5 thn Choos an optimal unforcd dg Rturn min{tsp5(g, F {}), tsp5(g {}, F )} Els /* thr is no rtx of dgr 5 in (G, F ) */ Rturn tsp4(g, F ). Not: Th input and output of algorithm tsp4(g, F ) ar as follows Input: An instanc (G, F ) such that th maximum dgr of G is at mos. Output: A mssag for th infasibility of (G, F ); or th minimum cost of a tour of (G, F ). Figur 3: Algorithm tsp5(g, F ) stting F := F {}, trmd as forc(), or by stting G := G {}, trmd dlt(). In fact, w bring to attntion that a rduction opration, if it dos not rturn a mssag Infasibl, is in fact a rpatd application of th abo two stps, forc() or dlt(), for som unforcd dg, idntifid by th conditions in Lmma 2. Thrfor, w procd with analyzing th ffcts of applying th forc() or th dlt() opration. Lt = u b an unforcd dg to which on of th forc() or dlt() oprations will b applid. Both u and must ha dgr mor than 2, othrwis by Lmma 1, th instanc is infasibl. Without loss of gnrality, w obsr th ffct of th opration on th rtx wight ω(). In th cas that opration forc() is applid, th following cass may aris. If is an unforcd rtx, thn will bcom forcd. By Eqs. (3) to (5), th wight ω() will not incras. If is a forcd rtx, thn ω() will bcom 0. If d F () = 2, thn by Lmma 1 th instanc will bcom infasibl. On th othr hand, if opration dlt() is applid, w obsr th following cass. If is ithr forcd or unforcd, and d() 3, thn th dgr of will dcras by on, and by Eqs. (6) and (7) ω() will not incras. If is ithr forcd or unforcd, and d() 2, thn by Lmma 1 th instanc will bcom infasibl. Following th abo obsrations, w conclud that th complt masur W (I) of a gin instanc I = (G, F ) of th forcd TSP will not incras as a rsult of th rduction and branching oprations of th algorithms in Figurs 1 and 3. To simplify som argumnts, w introduc th following notation: 3 = w 3 w 3, 4 = w 4 w 4, 5 = w 5 w 5, 4 3 = w 4 w 3, 5 4 = w 5 w 4, 5 3 = w 5 w 3, 4 3 = w 4 w 3, 5 4 = w 5 w 4, and 5 3 = w 5 w 3. To simplify th list of our branching ctors, w us th following notation: m 1 = min{w 3, w 3, 4 3, 4 3, 5 4, 5 4 }, (8) m 2 = min{w 3, 4 3, 4 3, 5 4, 5 4 }, (9) m 3 = min{w 3, 3, w 4, 4, w 5, 5 }, (10) m 4 = min{ 4 3, 4 3, 5 4, 5 4 }, (11) m 5 = min{w 4, w 4, 5 3, 5 3 }, (12) m 6 = min{ 4 3, 5 4, 5 4 }, (13) m 7 = min{ 5 4, 5 4 }, (14) m 8 = min{ 5 3, 5 3 }, (15) m 9 = min{w 3, w 3, 4 3, 4 3, 5 4 }, and (16) m 10 = min{w 3, 3, w 4, 4, 5 }. (17) 3 Rsults In ordr to adopt th masur and conqur mthod in our algorithm, w nd to st a masur W for a gin an instanc I = (G, F ) of th forcd TSP. To this ffct, w st a non-ngati rtx wight function ω : V R + in th graph G, and w us th sum of wights of all rtics in th graph as th masur W (I) of instanc I. That is, W (I) = (ω()). (18) V (G) W bring to attntion th fact that th numbr n of rtics in th graph G rmains unmodifid throughout th procss of th rduction and branching oprations. In addition to sking a propr masur, w also rquir that th wight of ach rtx is not gratr than 1, and thrfor, th masur W (I) will not b gratr than th numbr n of rtics in G. As a consqunc, a 2015 ISORA IET 50 Luoyang, China, Augus1 24, 2015

7 running tim bound as a function of th masur W (I) implis th sam running tim bound as a function of n. Th wight assignd to ach rtx typ plays an important rol, sinc th alu of th branching factor dpnds solly on ths wights. Lt th rtx wight function ω() b chosn as follows: w 3 = for an f3-rtx w 3 = for a u3-rtx w 4 = for an f4-rtx ω() = w 4 = for a u4-rtx (19) w 5 = for an f5-rtx w 5 = 1 for a u5-rtx 0 othrwis. Lmma 4 If th rtx wight function ω() is st as in Eq. (19), thn ach branching opration in Figur 3 has a branching factor not gratr than A proof of Lmma 4 will b drid analytically in th sral subsctions which follow. From th lmma, w gt our main rsult: Thorm 1 Th TSP in an n-rtx graph G with maximum dgr 5 can b sold in O ( n )-tim and polynomial-spac. In th rmaindr of th analysis, for an optimal dg =, w dnot N U () by {,,..., t a }, a = d U (), and N U ( ) \ {} by {t a+1, t a+1,..., t a+b }, b = d U ( ) 1. W assum without loss of gnrality that +i = t a+i for i = 1, 2,..., c, whr c = N U () N U ( ), th numbr of good nighbors that and ha in common. 3.1 Branching on Edgs Around f5- rtics (c-1 to c-9) This sction dris branching ctors for th branching opration on an optimal dg =, incidnt to an f5-rtx, distinguishing nin cass for conditions c-1 to c-9. Cas c-1: Thr xist rtics V f5 and N U (; V f3 ) such that N U () N U ( ) = (s Figur 4): W branch on dg. Not that N U ( ) \ {} = { }. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs,, and will b dltd from G by th rduction ruls. Both and will bcom rtics of dgr 2. From Eq. (19), th wight of rtics of dgr 2 is 0. So, th wight of rtx dcrass by w 5 and th wight of rtx dcrass by w 3. Each of th rtics, and can b ithr a typ f3, u3, f4, u4, f5, or u5-rtx, and ach of thir wights would dcras by at last m 1 = min { } w 3, w 3, 4 3, 4 3, 5 4, 5 4. If rtx t5 is (a) forc( ) in c-1 (b) dlt(t1) in c-1 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 4: Illustration of branching rul c-1, whr rtx V f5 and rtx N U (; V f3 ) such that N U () N U ( ) =. an f3-rtx (rsp., u3, f4, u4, f5, or a u5-rtx), thn th wight dcras α of rtx would b w 3 (rsp., w 3, 4 3, 4 3, 5 4, and 5 4 ). Thus, th total wight dcras in th branch of forc( ) is at last (w 5 + w 3 + 3m 1 + α). In th branch of dlt( ), dg will b dltd from G by th branching opration, and dg will b addd to F by th rduction ruls. Th wight of rtx dcrass by 5 4 and th wight of rtx dcrass by w 3. If rtx is an f3-rtx (rsp., u3, f4, u4, f5, or a u5-rtx), thn th wight dcras β of rtx would b w 3 (rsp., 3, w 4, 4, w 5, and 5 ). Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 4 + w 3 + β). As a rsult, w gt th following six branching ctors: (w 5 + w 3 + 3m 1 + α, w 5 w 4 + w 3 + β) (20) for (α, β) {(w 3, w 3 ), (w 3, 3 ), ( 4 3, w 4 ), ( 4 3, 4 ), ( 5 4, w 5 ), ( 5 4, 5 )}. c-2. Thr xist rtics V f5 and N U (; V f3 ) such that N U () N U ( ) = { } (s Figur 5): W branch on dg. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs,, and will b dltd from G by th rduction ruls. So, th wight of rtx dcrass by w 5, and th wight of rtx dcrass by w 3. Each of th rtics and can b ithr a typ f3, u3, f4, u4, f5, or u5-rtx, and ach of thir wights would dcras by at last m 1 = min { } w 3, w 3, 4 3, 4 3, 5 4, 5 4. Thr ar two possibl cass for th rtx typ of rtx. First, lt b an f3 or u3-rtx. Aftr prforming th branching opration, would bcom a rtx of dgr 1. By Lmma 1, cas (i), this is infasibl, and th algorithm will rturn a mssag of infasibility ISORA IET 51 Luoyang, China, Augus1 24, 2015

8 (a) forc( ) in c-2 (b) dlt( ) in c-2 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 5: Illustration of branching rul c-2, whr rtx V f5 and rtx N U (; V f3 ) such that N U () N U ( ) = { }. t 6 t 6 (a) forc( ) in c-3 (b) dlt( ) in c-3 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 6: Illustration of branching rul c-3, whr rtx V f5 and rtx N U (; V u3 ). Scond, lt b an f4, u4, f5, or u5-rtx. If is an f4-rtx (rsp., u4, f5, or a u5-rtx), thn th wight dcras α of rtx would b w 4 (rsp., w 4, 5 3, and 5 3 ). Thus, th total wight dcras in th branch of forc( ) is at last (w 5 +w 3 +2m 1 +α). In th branch of dlt( ), dg will b dltd from G by th branching opration, and dg will b addd to F by th rduction ruls. So, th wights of rtics and dcras by 5 4 and w 3, rspctily. If rtx is an f4-rtx (rsp., u4, f5, or a u5-rtx), thn th wight dcras β of rtx would b w 4 (rsp., 4, w 5, and 5 ). Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 4 + w 3 + β). As a rsult, w gt th following four branching ctors: (w 5 + w 3 + 2m 1 + α, w 5 w 4 + w 3 + β) (21) for (α, β) {(w 4, w 4 ), (w 4, 4 ), ( 5 3, w 5 ), ( 5 3, 5 )}. c-3. Thr xist rtics V f5 and N U (; V u3 ) (s Figur 6): W branch on dg. Not that N U ( ) \ {} = {, t 6 }. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs, and will b dltd from G by th rduction ruls. So, th wights of rtics and dcras by w 5 and 3, rspctily. Each of rtics, and can b ithr a typ u3, f4, u4, f5, or u5-rtx, and ach of thir wights would dcras by at last m 2 = min { w 3, 4 3, 4 3, 5 4, 5 4 }. Thus, th total wight dcras in th branch of forc( ) is at last (w 5 + w 3 w 3 + 3m 2 ). In th branch of dlt( ), dg will b dltd from G by th branching opration and dgs and t 6 will b addd to F by th rduction ruls. So, th wight of rtx dcrass by 5 4 and th wight of rtx dcrass by w 3. Each of th rtics and t 6 can b ithr a typ f3, u3, f4, u4, f5, or u5-rtx, and ach of thir wights would dcras by at last m 3 = min {w 3, 3, w 4, 4, w 5, 5 }. Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 4 + w 3 + 2m 3 ). As a rsult, w gt th following branching ctor: (w 5 + w 3 w 3 + 3m 2, w 5 w 4 + w 3 + 2m 3 ). (22) c-4. Thr xist rtics V f5 and N U (; V f4 ) such that N U () N U ( ) = (s Figur 7): W branch on dg. Not that N U ( ) \ {} = {, t 6 }. t 6 (a) forc( ) in c-4 t 6 (b) dlt( ) in c-4 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 7: Illustration of branching rul c-4, whr rtx V f5 and rtx N U (; V f4 ), such that N U () N U ( ) =. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs,,, and t 6 will b dltd from G by th rduction ruls. So, th wight of rtx dcrass by w 5 and th wight of rtx dcrass by w 4. Each of th rtics, and can b ithr a typ f4, u4, f5, or u5-rtx, and ach of thir wights would dcras by at last m 4 = min { } 4 3, 4 3, 5 4, 5 4. Each of rtics t5 and t 6 can b ithr a typ f3, u3, f4, u4, f5, or u5- rtx, and ach of thir wights would dcras by 2015 ISORA IET 52 Luoyang, China, Augus1 24, 2015

9 at last m 1 = min { w 3, w 3, 4 3, 4 3, 5 4, 5 4 }. Thus, th total dcras in th branch of forc( ) is at last (w 5 + w 4 + 3m 4 + 2m 1 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, th wight of rtx dcrass by 5 4 and th wight of rtx dcrass by 4 3. Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 3 ). As a rsult, w gt th following branching ctor: (w 5 + w 4 + 3m 4 + 2m 1, w 5 w 3 ). (23) c-5. Thr xist rtics V f5 and N U (; V f4 ) such that N U () N U ( ). W distinguish two sub cass, according to th cardinality of th intrsction N U () N U ( ), (c-5(i)), N U () N U ( ) = 1, and (c-5(ii)), N U () N U ( ) = 2. c-5(i). Without loss of gnrality, assum that N U () N U ( ) = { } (s Figur 8): W branch on dg. Not that N U ( ) \ {} = { }. (a) forc(t1) in c-5(i) (b) dlt(t1) in c-5(i) : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 8: Illustration of branching rul c-5(i), whr rtx V f5 and rtx N U (; V f4 ), such that N U () N U ( ) = { }. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs,,,, and will b dltd from G by th rduction ruls. So, th wight of rtx dcrass by w 5 and th wight of rtx dcrass by w 4. Vrtx can b ithr a typ f4, u4, f5, or u5-rtx, and its wight would dcras by at last m 5 = min { } w 4, w 4, 5 3, 5 3. Each of th rtics and can b ithr a typ f4, u4, f5, or u5- rtx, and ach of thir wights would dcras by at last m 4 = min { } 4 3, 4 3, 5 4, 5 4. Vrtx can b ithr a typ f3, u3, f4, u4, f5, or u5-rtx, and its wight would dcras by at last m 1 = min { } w 3, w 3, 4 3, 4 3, 5 4, 5 4. Thus, th total wight dcras in th branch of forc( ) is at last (w 5 + w 4 + m 5 + 2m 4 + m 1 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, th wight of rtx dcrass by 5 4, and th wight of rtx dcrass by 4 3. Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 3 ). As a rsult, w gt th following branching ctor: (w 5 + w 4 + m 5 + 2m 4 + m 1, w 5 w 3 ). (24) c-5(ii). Without loss of gnrality, assum that N U () N U ( ) = {, } (s Figur 9): W branch on dg. (a) forc( ) in c-5(ii) (b) dlt( ) in c-5(ii) : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 9: Illustration of branching rul c-5(ii), whr rtx V f5 and rtx N U (; V f4 ) such that N U () N U ( ) = {, }. In th branch of forc( ), dg will b addd to F by th branching opration, and dg,,, and will b dltd from G by th rduction ruls. So, th wight of rtx dcrass by w 5 and th wight of rtx dcrass by w 4. Each of rtics and can b ithr a typ f4, u4, f5, or u5-rtx, and ach of thir wights would dcras by at last m 5 = min { } w 4, w 4, 5 3, 5 3. Vrtx can b ithr a typ f3, u3, f4, u4, f5, or u5-rtx, and its wight would dcrass by at last m 4 = min { } 4 3, 4 3, 5 4, 5 4. Thus, th total wight dcras in th branch of forc( ) is at last (w 5 + w 4 + 2m 5 + m 4 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, th wight of rtx dcrass by 5 4, and th wight of rtx dcrass by 4 3. Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 3 ). As a rsult, w gt th following branching ctor: (w 5 + w 4 + 2m 5 + m 4, w 5 w 3 ). (25) c-6. Thr xist rtics V f5 and N U (; V u4 ) (s Figur 10): W branch on dg. Not that N U ( ) \ {} = {, t 6, t 7 }. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs, and will b dltd from G by th rduction ruls. So, th wight of rtx dcrass by w 5 and th wight of rtx dcrass by 4. Each of th rtics, and can b ithr a typ u4, f5, or u5- rtx, and ach of thir wights would dcras by at 2015 ISORA IET 53 Luoyang, China, Augus1 24, 2015

10 t6 t 7 (a) forc( ) in c-6 (b) dlt( ) in c-6 : unforcd dgs : forcd dgs t6 t 7 : nwly dltd dgs : nwly forcd dgs Figur 10: Illustration of branching rul c-6, whr rtx V f5 and rtx N U (; V u4 ). last m 6 = min { 4 3, 5 4, 5 4 }. Thus, th total wight dcras in th branch of forc( ) is at last (w 5 + w 4 w 4 + 3m 6 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, th wight of rtx dcrass by 5 4 and th wight of rtx dcrass by 4 3. Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 4 + w 4 w 3 ). As a rsult, w gt th following branching ctor: (w 5 + w 4 w 4 + 3m 6, w 5 w 4 + w 4 w 3 ). (26) c-7. Thr xist rtics V f5 and N U (; V f5 ) such that N U () N U ( ) = (s Figur 11): W branch on dg. Not that N U ( ) \ {} = {, t 6, t 7 }. t6 t 7 (a) forc( ) in c-7 t6 t 7 (b) dlt( ) in c-7 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 11: Illustration of branching rul c-7, whr rtx V f5 and rtx N U (; V f5 ), such that N U () N U ( ) =. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs,,,, t 6 and t 7 will b dltd from G by th rduction ruls. So, both wights of rtx and rtx dcrass by w 5, ach. Each of rtics,,,, t 6 and t 7 can b ithr a typ f5, or u5- rtx, and ach of thir wights would dcras by at last m 7 = min { 5 4, 5 4 }. Thus, th total wight dcras in th branch of forc( ) is at last (2w 5 + 6m 7 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, both wights of rtics and dcrass by 5 4, ach. Thus, th total wight dcras in th branch of dlt( ) is at last (2w 5 2w 4 ). As a rsult, w gt th following branching ctor: (2w 5 + 6m 7, 2w 5 2w 4 ). (27) c-8. Thr xist rtics V f5 and N U (; V f5 ) such that N U () N U ( ). W distinguish thr sub cass, according to th cardinality of th intrsction N U () N U ( ), (c-8(i)), N U () N U ( ) = 1, (c-8(ii)), N U () N U ( ) = 2, and (c-8(iii)), N U () N U ( ) = 3. c-8(i). Without loss of gnrality, assum that N U () N U ( ) = { } (s Figur 12): W branch on dg. Not that N U ( ) \ {} = {, t 6 }. t 6 (a) forc( ) in c-8(i) t 6 (b) dlt( ) in c-8(i) : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 12: Illustration of branching rul c-8(i), whr rtx V f5 and rtx N U (; V f5 ), such that N U () N U ( ) = { }. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs,,,, and t 6 will b dltd from G by th rduction ruls. Both wights of rtx and rtx dcrass by w 5, ach. Vrtx can b ithr a typ f5 or u5-rtx, and its wight would dcrass by at last m 8 = min { 5 3, 5 3 }. Each of rtics t3,, and t 6 can b ithr a typ f5, or u5-rtx, and ach of thir wights would dcras by at last m 7 = min { 5 4, 5 4 }. Thus, th total wight dcras in th branch of forc( ) is at last (2w 5 + 4m 7 + m 8 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. Both wights of rtics and dcrass by 5 4, ach. Thus, th total wight dcras in th branch of dlt( ) is at last (2w 5 2w 4 ). As a rsult, w gt th following branching ctor: (2w 5 + 4m 7 + m 8, 2w 5 2w 4 ). (28) 2015 ISORA IET 54 Luoyang, China, Augus1 24, 2015

11 c-8(ii). Without loss of gnrality, assum that N U () N U ( ) = {, } (s Figur 13): W branch on dg. Not that N U ( ) \ {} = { }. (a) forc( ) in c-8(ii) (b) dlt( ) in c-8(ii) : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 13: Illustration of branching rul c-8(ii), whr rtx V f5 and rtx N U (; V f5 ), such that N U () N U ( ) = {, }. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs,,,, and will b dltd from G by th rduction ruls. So, both wights of rtx and rtx dcrass by w 5, ach. Each of rtics and can b ithr a typ f5, or u5-rtx, and ach of thir wights would dcras by at last m 8 = min { } 5 3, 5 3. Each of rtics t4 and can b ithr a typ f5, or u5-rtx, and ach of thir wights would dcras by at last m 7 = min { } 5 4, 5 4. Thus, th total wight dcras in th branch of forc( ) is at last (2w 5 +2m 8 +2m 7 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, both wights of rtx and rtx dcrass by 5 4, ach. Th total wight dcras in th branch of dlt( ) is at last (2w 5 2w 4 ). As a rsult, w gt th following branching ctor: (2w 5 + 2m 8 + 2m 7, 2w 5 2w 4 ). (29) c-8(iii). Without loss of gnrality, assum that N U () N U ( ) = {,, } (s Figur 14): W branch on dg. In th branch of forc( ), dg will b addd to F by th branching opration, and dgs,,,, and will b dltd from G by th rduction ruls. So, both wights of rtx and rtx dcrass by w 5, ach. Each of rtics,, and can b ithr a typ f5, or u5- rtx, and ach of thir wights would dcras by at last m 8 = min { } 5 3, 5 3. Thus, th total wight dcras in th branch of forc( ) is at last (2w 5 + 3m 8 ). (a) forc( ) in c-8(iii) (b) dlt( ) in c-8(iii) : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 14: Illustration of branching rul c-8(iii), whr rtx V f5 and rtx N U (; V f5 ), such that N U () N U ( ) = {,, }. In th branch of dlt( ), dg will b dltd from G by th branching opration. Thus, both wights of rtx and rtx dcrass by 5 4, ach. Th total wight dcras in th branch of dlt( ) is at last (2w 5 2w 4 ). As a rsult, w gt th following branching ctor: (2w 5 + 3m 8, 2w 5 2w 4 ). (30) c-9. Thr xist rtics V f5 and N U (; V u5 ) (s Figur 15): W branch on dg. t 8 t6 t 7 (a) forc( ) in c-9 t 8 t6 t 7 (b) dlt( ) in c-9 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 15: Illustration of branching rul c-9, whr rtx V f5 and rtx N U (; V u5 ). In th branch of forc( ), dg will b addd to F by th branching opration, and dgs, and will b dltd from G by th rduction ruls. So, th wight of rtx dcrass by w 5, and th wight of rtx dcrass by 5. Each of rtics, and can only b a typ u5-rtx, and ach of thir wights dcras by 5 4. Thus, th total wight dcras in th branch of forc( ) is at last (4w 5 3w 4 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. Thus, th wight of rtx dcrass by 5 4, and th wight of rtx dcrass by 5 4. Th total wight dcras in 2015 ISORA IET 55 Luoyang, China, Augus1 24, 2015

12 th branch of dlt( ) is at last (w 5 + w 5 w 4 w 4 ). Thn, w gt th following branching ctor: N U (; V f4 ) (s Figur 17): W branch on dg. Not that N U ( ) \ {} = {t 6, t 7 }. (4 3w 4, 1 + w 5 w 4 w 4 ). (31) 3.2 Branching on Edgs Around u5- rtics (c-10 to c-14) If non of th first nin conditions can b xcutd, this mans that th graph has no f5-rtics. But this dos not man that th maximum dgr of th graph has bn rducd to 4, sinc thr might still b u5-rtics. This sction dris branching ctors for branchings on an optimal dg = incidnt to a u5-rtx, distinguishing th fi cass for conditions c-10 to c-14. c-10. Thr xist rtics V u5 and N U (; V f3 ) (s Figur 16): W branch on dg. Not that N U ( ) \ {} = {t 6 }. t 6 (a) forc( ) in c-10 (b) dlt( ) in c-10 t 6 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 16: Illustration for c-10 whr rtics V u5 and N U (; V f3 ). In th branch of forc( ), dg will b addd to F by th branching opration, and dg t 6 will b dltd from G by th rduction ruls. So, th wight of rtx dcrass by 5, and th wight of rtx dcrass by w 3. If rtx t 6 is an f3-rtx (rsp., u3, f4, u4, or a u5-rtx), thn th wight dcras α of rtx t 6 would b w 3 (rsp., w 3, 4 3, 4 3, and 5 4 ). Thus, th total wight dcras in th branch of forc( ) is at last (w 5 w 5 + w 3 + α). In th branch of dlt( ), dg will b dltd from G by th branching opration, and dg t 6 will b addd to F by th rduction ruls. Th wight of rtx dcrass by 5 4, and th wight of rtx dcrass by w 3. If rtx t 6 is an f3-rtx (rsp., u3, f4, u4, or a u5-rtx), thn th wight dcras β of rtx t 6 would b w 3 (rsp., 3, w 4, 4, and 5 ). Thus, total wight dcras in th branch of dlt( ) is at last (w 5 w 4 + w 3 + β). As a rsult, w gt fi branching ctors: (1 w 5 + w 3 + α, 1 w 4 + w 3 + β) (32) for (α, β) {(w 3, w 3 ), (w 3, 3 ), ( 4 3, w 4 ), ( 4 3, 4 ), ( 5 4, 5 )}. c-11. Thr xist rtics V u5 and t 6 t 7 t 6 t 7 (a) forc( ) in c-11 (b) dlt( ) in c-11 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 17: Illustration for c-11 whr rtics V u5 and N U (; V f4 ). In th branch of forc( ), dg will b addd to F by th branching opration, and dgs t 6 and t 7 will b dltd from G by th rduction ruls. So, th wight of rtx dcrass by 5, and th wight of rtx dcrass by w 4. Each of rtics t 6 and t 7 can b ithr a typ f3, u3, f4, u4, or u5- rtx, and ach of thir wights would dcras by at last m 9 = min { w 3, w 3, 4 3, 4 3, 5 4 }. Thus, th total wight dcras in th branch of forc( ) is at last (w 5 w 5 + w 4 + 2m 9 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, th wight of rtx dcrass by 5 4, and th wight of rtx dcrass by 4 3. Th total wight dcras in th branch of dlt( ) is at last (w 5 w 4 + w 4 w 3 ). As a rsult, w gt th following branching ctor: (1 w 5 + w 4 + 2m 9, 1 w 4 + w 4 w 3 ). (33) c-12. Thr xist rtics V u5 and t N U (; V u3 ) (s Figur 18): W branch on dg. Not that N U ( ) \ {} = {t 6, t 7 }. t 6 t 7 (a) forc( ) in c-12 (b) dlt( ) in c-12 t 6 t 7 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 18: Illustration of branching rul c-12, whr rtx V u5 and rtx t N U (; V u3 ). In th branch of forc( ), dg will b addd to F by th branching opration. So, th wight of rtx dcrass by 5, and th wight of rtx dcrass by 3. Th total wight dcras in th 2015 ISORA IET 56 Luoyang, China, Augus1 24, 2015

13 branch of forc( ) is at last (w 5 w 5 + w 3 w 3 ). In th branch of dlt( ), dg will b dltd from G by th branching opration, and dgs t 6 and t 7 will b addd to F by th rduction ruls. So, th wight of rtx dcrass by 5 4, and th wight of rtx dcrass by w 3. Each of rtics t 6 and t 7 can b ithr a typ f3, u3, f4, u4, or u5-rtx, and ach of thir wights would dcras by at last m 10 = min{w 3, 3, w 4, 4, 5 }. Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 4 + w 3 + 2m 10 ). As a rsult, w gt th following branching ctor: (1 w 5 + w 3 w 3, 1 w 4 + w 3 + 2m 10 ). (34) c-13. Thr xist rtics V u5 and N U (; V u4 ) (s Figur 19): W branch on dg. t 6 t7 t 8 (a) forc( ) in c-13 (b) dlt( ) in c-13 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 19: Illustration of branching rul c-13, whr rtx V u5 and rtx N U (; V u4 ). In th branch of forc( ), dg will b addd to F by th branching opration. So, th wight of rtx dcrass by 5, and th wight of rtx dcrass by 4. Thus, th total wight dcras in th branch of forc( ) is at last (w 5 w 5 + w 4 w 4 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, th wight of rtx dcrass by 5 4, and th wight of rtx dcrass by 4 3. Thus, th total wight dcras in th branch of dlt( ) is at last (w 5 w 3 ). Thn, w gt th following branching ctor: t 6 (1 w 5 + w 4 w 4, 1 w 3 ). (35) c-14. Thr xist rtics V u5 and N U (; V u5 ) (s Figur 20): W branch on dg. In th branch of forc( ), dg will b addd to F by th branching opration. So, both wights of rtx and rtx dcrass by 5, ach. Thus, th total wight dcras in th branch of forc( ) is at last (2w 5 2w 5 ). In th branch of dlt( ), dg will b dltd from G by th branching opration. So, both wights of rtx and rtx dcrass by 5 4, ach. Thus, th total wight dcras in th branch of dlt( ) is at last (2w 5 2w 4 ). Thn, t7 t 8 t t 6 9 t 7 t 8 t t 6 9 t 7 t 8 (a) forc( ) in c-14 (b) dlt( ) in c-14 : unforcd dgs : forcd dgs : nwly dltd dgs : nwly forcd dgs Figur 20: Illustration of branching rul c-14, whr rtx V u5 and rtx N U (; V u5 ). w gt th following branching ctor: 3.3 Switching to TSP4 (2 2w 5, 2 2w 4 ). (36) If non of ths 14 cass can b xcutd, this mans that th graph has no mor dgr-5 rtics. In that cas, w can switch and us a fast algorithm for TSP in dgr-4 graphs (tsp4(g, F )) to sol th rmaining instancs. Xiao and Nagamochi [15, Lmma 3] ha shown how to lrag rsults obtaind by a masurand-conqur analysis, and that an algorithm can b usd as a subprocdur, gin that w know th rspcti wight stting mchanism. To gt a combination of total running tim bound of ths two algorithms, w can us th maximum branching factor for TSP in dgr-4 graphs algorithm and a masur µ is calculatd basd on th maximum ratio of rtx wights for TSP in dgr-4 graphs and TSP in dgr- 5 graphs [12]. Hr w us th O ( n )-tim algorithm by Xiao and Nagamochi [13], whr th wights of rtics in dgr-4 graphs ar st as follows; w 3 = , w 3 = , w 4 = , and w 4 = 1. For this stp, th running tim bound is { }) T (µ) O ( max w, w, w, 1 4 w Orall Analysis (37) Th branching factor of ach of th branching ctors from (20) to (37) dos not xcd Th tight constraints in th quasiconx program ar in conditions c-4, c-10, c-11, c-12, c-13 and th switching constraint. This complts a proof of Thorm 1. 4 Conclusion In this papr, w ha prsntd an xact algorithm for TSP in dgr-5 graphs. Our algorithm is a simpl 2015 ISORA IET 57 Luoyang, China, Augus1 24, 2015

14 branching algorithm, following th branch-and-rduc paradigm, and it oprats in spac which is polynomial of th siz of an input instanc. To th bst of our knowldg, this is th first polynomial spac xact algorithm dlopd spcifically for graphs of maximum dgr at mos, and xtnds prious algorithms for dgr 3 [11, 14], and dgr-4 graphs [13]. W usd th masur and conqur mthod for th analysis of th running tim of th proposd algorithm, and ha obtaind an uppr bound of O ( n ), whr n is th numbr of rtics in a gin instanc. This rsult compars faorably with th polynomialspac TSP algorithm for gnral graphs by Gurich and Shlah [8], which runs in O (4 n n log n )-tim. It rmains an opn qustion whthr this tim bound can b furthr improd by a modifid analysis tchniqu, or by a carful r-xamination of th branching ruls. Indd, it would b most intrsting to obtain a polynomial-spac algorithm with a running tim of O (2 n ) or lss, or simply show that this cannot b achid. Acknowldgmnts Th first author would lik to xprss gratitud to Tchnical Unirsity of Malaysia Malacca, Malaysia and Ministry of Highr Education (MOHE) Malaysia for th scholarship program. Rfrncs [1] Bllman, R. : Combinatorial Procsss and Dynamic Programming. In: Procding of th 10th Symposium in Applid Mathmatics, Amr. Math. Soc., Proidnc, RI, [2] Bodlandr, H. L., Cygan M., Kratsch S. and Ndrlof J. : Soling Wightd and Counting Variants of Connctiity Problms Paramtrizd by Trwidth Dtrministically in Singl Exponntial Tim. In: CoRR abs/ , [3] Eppstin, D. : Quasiconx Analysis of Multiariat Rcurrnc Equations for Backtracking Algorithms. In: ACM Transactions on Algorithms, Vol. 2, No. 4, pp , [4] Eppstin, D. : Th Traling Salsman Problm for Cubic Graphs. In: Journal of Graph Algorithms and Application, Vol. 11, No. 1, pp , [5] Fomin, F. V., Grandoni, F. and Kratsch, D. : A Masur and Conqur Approach for th Analysis of Exact Algorithms. In: J. ACM, Vol. 56, No. 5, Articl 25, [6] Fomin, F. V. and Kratsch, D. : Exact Exponntial Algorithms. In: Brlin Hidlbrg: Springr, [7] Gbaur, H. : Finding and Enumrating Hamilton Cycls in 4-rgular Graphs. In: Thortical Computr Scinc, Vol. 412, No. 35, pp , [8] Gurich, Y. and Shlah, S. : Expctd Computation Tim for Hamiltonian Path Problm. In: Siam Journal of Computation, Vol. 16, No. 3, pp , [9] Hld, M. and Karp, R. M. : A Dynamic Programming Approach to Squncing Problms. In: Journal of th Socity for Industrial and Applid Mathmatics, Vol. 10, No. 1, pp , [10] Iwama, K. and Nakashima, T. : An Improd Exact Algorithm for Cubic Graph TSP. In: CO- COON, LNCS 4598, pp , [11] Liskiwicz, M. and Schustr, M. R. : A Nw Uppr Bound for th Traling Salsman Problm in Cubic Graphs. In: CoRR abs/ , [12] Xiao, M. and Nagamochi, H. : Furthr Impromnt on Maximum Indpndnt St in Dgr-4 Graphs. In: COCOA 2011, LNCS 6831, pp , [13] Xiao, M. and Nagamochi, H. : An Improd Exact Algorithm for TSP in Graphs of Maximum Dgr-4. In: Thory Comput Syst, DOI /s x, [14] Xiao, M. and Nagamochi, H. : An Exact Algorithm for TSP in Dgr-3 Graphs ia Circuit Procdur and Amortization on Connctiity Structur. In: TAMC 2013, LNCS 7876, pp , [15] Xiao, M. and Nagamochi, H. : Exact Algorithms for Maximum Indpndnt St. In: ISAAC LNCS 8283, pp , ISORA IET 58 Luoyang, China, Augus1 24, 2015

(Upside-Down o Direct Rotation) β - Numbers

(Upside-Down o Direct Rotation) β - Numbers Amrican Journal of Mathmatics and Statistics 014, 4(): 58-64 DOI: 10593/jajms0140400 (Upsid-Down o Dirct Rotation) β - Numbrs Ammar Sddiq Mahmood 1, Shukriyah Sabir Ali,* 1 Dpartmnt of Mathmatics, Collg

More information

1 Minimum Cut Problem

1 Minimum Cut Problem CS 6 Lctur 6 Min Cut and argr s Algorithm Scribs: Png Hui How (05), Virginia Dat: May 4, 06 Minimum Cut Problm Today, w introduc th minimum cut problm. This problm has many motivations, on of which coms

More information

Some Results on E - Cordial Graphs

Some Results on E - Cordial Graphs Intrnational Journal of Mathmatics Trnds and Tchnology Volum 7 Numbr 2 March 24 Som Rsults on E - Cordial Graphs S.Vnkatsh, Jamal Salah 2, G.Sthuraman 3 Corrsponding author, Dpartmnt of Basic Scincs, Collg

More information

A Polynomial-Space Exact Algorithm for TSP in Degree-8 Graphs

A Polynomial-Space Exact Algorithm for TSP in Degree-8 Graphs A Polynomial-Spac Exac Algorihm for TSP in Dgr-8 Graphs Norhazwani Md Yunos Alksandar Shurbski Hiroshi Nagamochi Dparmn of Applid Mahmaics and Physics, Gradua School of Informaics, Kyoo Unirsiy {wani,

More information

Vishnu V. Narayan. January

Vishnu V. Narayan. January A 17 12 -approimation algorithm for 2-rt-connctd spanning subgraphs on graphs with minimum dgr at last arxi:1612.047902 [cs.ds] 17 Jan 2017 Vishnu V. Naraan Januar 17 2017 W obtain a polnomial-tim 17 -approimation

More information

Finding low cost TSP and 2-matching solutions using certain half integer subtour vertices

Finding low cost TSP and 2-matching solutions using certain half integer subtour vertices Finding low cost TSP and 2-matching solutions using crtain half intgr subtour vrtics Sylvia Boyd and Robrt Carr Novmbr 996 Introduction Givn th complt graph K n = (V, E) on n nods with dg costs c R E,

More information

The Equitable Dominating Graph

The Equitable Dominating Graph Intrnational Journal of Enginring Rsarch and Tchnology. ISSN 0974-3154 Volum 8, Numbr 1 (015), pp. 35-4 Intrnational Rsarch Publication Hous http://www.irphous.com Th Equitabl Dominating Graph P.N. Vinay

More information

On spanning trees and cycles of multicolored point sets with few intersections

On spanning trees and cycles of multicolored point sets with few intersections On spanning trs and cycls of multicolord point sts with fw intrsctions M. Kano, C. Mrino, and J. Urrutia April, 00 Abstract Lt P 1,..., P k b a collction of disjoint point sts in R in gnral position. W

More information

Differential Equations

Differential Equations UNIT I Diffrntial Equations.0 INTRODUCTION W li in a world of intrrlatd changing ntitis. Th locit of a falling bod changs with distanc, th position of th arth changs with tim, th ara of a circl changs

More information

Basic Polyhedral theory

Basic Polyhedral theory Basic Polyhdral thory Th st P = { A b} is calld a polyhdron. Lmma 1. Eithr th systm A = b, b 0, 0 has a solution or thr is a vctorπ such that π A 0, πb < 0 Thr cass, if solution in top row dos not ist

More information

Propositional Logic. Combinatorial Problem Solving (CPS) Albert Oliveras Enric Rodríguez-Carbonell. May 17, 2018

Propositional Logic. Combinatorial Problem Solving (CPS) Albert Oliveras Enric Rodríguez-Carbonell. May 17, 2018 Propositional Logic Combinatorial Problm Solving (CPS) Albrt Olivras Enric Rodríguz-Carbonll May 17, 2018 Ovrviw of th sssion Dfinition of Propositional Logic Gnral Concpts in Logic Rduction to SAT CNFs

More information

Searching Linked Lists. Perfect Skip List. Building a Skip List. Skip List Analysis (1) Assume the list is sorted, but is stored in a linked list.

Searching Linked Lists. Perfect Skip List. Building a Skip List. Skip List Analysis (1) Assume the list is sorted, but is stored in a linked list. 3 3 4 8 6 3 3 4 8 6 3 3 4 8 6 () (d) 3 Sarching Linkd Lists Sarching Linkd Lists Sarching Linkd Lists ssum th list is sortd, but is stord in a linkd list. an w us binary sarch? omparisons? Work? What if

More information

Construction of asymmetric orthogonal arrays of strength three via a replacement method

Construction of asymmetric orthogonal arrays of strength three via a replacement method isid/ms/26/2 Fbruary, 26 http://www.isid.ac.in/ statmath/indx.php?modul=prprint Construction of asymmtric orthogonal arrays of strngth thr via a rplacmnt mthod Tian-fang Zhang, Qiaoling Dng and Alok Dy

More information

Homework #3. 1 x. dx. It therefore follows that a sum of the

Homework #3. 1 x. dx. It therefore follows that a sum of the Danil Cannon CS 62 / Luan March 5, 2009 Homwork # 1. Th natural logarithm is dfind by ln n = n 1 dx. It thrfor follows that a sum of th 1 x sam addnd ovr th sam intrval should b both asymptotically uppr-

More information

Week 3: Connected Subgraphs

Week 3: Connected Subgraphs Wk 3: Connctd Subgraphs Sptmbr 19, 2016 1 Connctd Graphs Path, Distanc: A path from a vrtx x to a vrtx y in a graph G is rfrrd to an xy-path. Lt X, Y V (G). An (X, Y )-path is an xy-path with x X and y

More information

Search sequence databases 3 10/25/2016

Search sequence databases 3 10/25/2016 Sarch squnc databass 3 10/25/2016 Etrm valu distribution Ø Suppos X is a random variabl with probability dnsity function p(, w sampl a larg numbr S of indpndnt valus of X from this distribution for an

More information

Mutually Independent Hamiltonian Cycles of Pancake Networks

Mutually Independent Hamiltonian Cycles of Pancake Networks Mutually Indpndnt Hamiltonian Cycls of Pancak Ntworks Chng-Kuan Lin Dpartmnt of Mathmatics National Cntral Univrsity, Chung-Li, Taiwan 00, R O C discipl@ms0urlcomtw Hua-Min Huang Dpartmnt of Mathmatics

More information

BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES

BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES DONALD M. DAVIS Abstract. If p is a prim (implicit in notation and n a positiv intgr, lt ν(n dnot th xponnt of p in n, and U(n n/p ν(n, th unit

More information

CPSC 665 : An Algorithmist s Toolkit Lecture 4 : 21 Jan Linear Programming

CPSC 665 : An Algorithmist s Toolkit Lecture 4 : 21 Jan Linear Programming CPSC 665 : An Algorithmist s Toolkit Lctur 4 : 21 Jan 2015 Lcturr: Sushant Sachdva Linar Programming Scrib: Rasmus Kyng 1. Introduction An optimization problm rquirs us to find th minimum or maximum) of

More information

LINEAR DELAY DIFFERENTIAL EQUATION WITH A POSITIVE AND A NEGATIVE TERM

LINEAR DELAY DIFFERENTIAL EQUATION WITH A POSITIVE AND A NEGATIVE TERM Elctronic Journal of Diffrntial Equations, Vol. 2003(2003), No. 92, pp. 1 6. ISSN: 1072-6691. URL: http://jd.math.swt.du or http://jd.math.unt.du ftp jd.math.swt.du (login: ftp) LINEAR DELAY DIFFERENTIAL

More information

cycle that does not cross any edges (including its own), then it has at least

cycle that does not cross any edges (including its own), then it has at least W prov th following thorm: Thorm If a K n is drawn in th plan in such a way that it has a hamiltonian cycl that dos not cross any dgs (including its own, thn it has at last n ( 4 48 π + O(n crossings Th

More information

Approximation and Inapproximation for The Influence Maximization Problem in Social Networks under Deterministic Linear Threshold Model

Approximation and Inapproximation for The Influence Maximization Problem in Social Networks under Deterministic Linear Threshold Model 20 3st Intrnational Confrnc on Distributd Computing Systms Workshops Approximation and Inapproximation for Th Influnc Maximization Problm in Social Ntworks undr Dtrministic Linar Thrshold Modl Zaixin Lu,

More information

GRAPH THEORY AND APPLICATIONS 2 MARKS QUESTIONS AND ANSWERS

GRAPH THEORY AND APPLICATIONS 2 MARKS QUESTIONS AND ANSWERS GRAPH THEORY AND APPLICATIONS MARKS QUESTIONS AND ANSWERS UNIT I INTRODUCTION 1. Dfin Graph. A graph G = (V, E) consists of a st of objcts V={1,, 3, } calld rtics (also calld points or nods) and othr st

More information

Application of Vague Soft Sets in students evaluation

Application of Vague Soft Sets in students evaluation Availabl onlin at www.plagiarsarchlibrary.com Advancs in Applid Scinc Rsarch, 0, (6):48-43 ISSN: 0976-860 CODEN (USA): AASRFC Application of Vagu Soft Sts in studnts valuation B. Chtia*and P. K. Das Dpartmnt

More information

Section 6.1. Question: 2. Let H be a subgroup of a group G. Then H operates on G by left multiplication. Describe the orbits for this operation.

Section 6.1. Question: 2. Let H be a subgroup of a group G. Then H operates on G by left multiplication. Describe the orbits for this operation. MAT 444 H Barclo Spring 004 Homwork 6 Solutions Sction 6 Lt H b a subgroup of a group G Thn H oprats on G by lft multiplication Dscrib th orbits for this opration Th orbits of G ar th right costs of H

More information

Higher order derivatives

Higher order derivatives Robrto s Nots on Diffrntial Calculus Chaptr 4: Basic diffrntiation ruls Sction 7 Highr ordr drivativs What you nd to know alrady: Basic diffrntiation ruls. What you can larn hr: How to rpat th procss of

More information

NEW APPLICATIONS OF THE ABEL-LIOUVILLE FORMULA

NEW APPLICATIONS OF THE ABEL-LIOUVILLE FORMULA NE APPLICATIONS OF THE ABEL-LIOUVILLE FORMULA Mirca I CÎRNU Ph Dp o Mathmatics III Faculty o Applid Scincs Univrsity Polithnica o Bucharst Cirnumirca @yahoocom Abstract In a rcnt papr [] 5 th indinit intgrals

More information

Strongly Connected Components

Strongly Connected Components Strongly Connctd Componnts Lt G = (V, E) b a dirctd graph Writ if thr is a path from to in G Writ if and is an quivalnc rlation: implis and implis s quivalnc classs ar calld th strongly connctd componnts

More information

BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES. 1. Statement of results

BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES. 1. Statement of results BINOMIAL COEFFICIENTS INVOLVING INFINITE POWERS OF PRIMES DONALD M. DAVIS Abstract. If p is a prim and n a positiv intgr, lt ν p (n dnot th xponnt of p in n, and u p (n n/p νp(n th unit part of n. If α

More information

ARIMA Methods of Detecting Outliers in Time Series Periodic Processes

ARIMA Methods of Detecting Outliers in Time Series Periodic Processes Articl Intrnational Journal of Modrn Mathmatical Scincs 014 11(1): 40-48 Intrnational Journal of Modrn Mathmatical Scincs Journal hompag:www.modrnscintificprss.com/journals/ijmms.aspx ISSN:166-86X Florida

More information

Homotopy perturbation technique

Homotopy perturbation technique Comput. Mthods Appl. Mch. Engrg. 178 (1999) 257±262 www.lsvir.com/locat/cma Homotopy prturbation tchniqu Ji-Huan H 1 Shanghai Univrsity, Shanghai Institut of Applid Mathmatics and Mchanics, Shanghai 272,

More information

Introduction to Arithmetic Geometry Fall 2013 Lecture #20 11/14/2013

Introduction to Arithmetic Geometry Fall 2013 Lecture #20 11/14/2013 18.782 Introduction to Arithmtic Gomtry Fall 2013 Lctur #20 11/14/2013 20.1 Dgr thorm for morphisms of curvs Lt us rstat th thorm givn at th nd of th last lctur, which w will now prov. Thorm 20.1. Lt φ:

More information

The Matrix Exponential

The Matrix Exponential Th Matrix Exponntial (with xrciss) by D. Klain Vrsion 207.0.05 Corrctions and commnts ar wlcom. Th Matrix Exponntial For ach n n complx matrix A, dfin th xponntial of A to b th matrix A A k I + A + k!

More information

SOME PARAMETERS ON EQUITABLE COLORING OF PRISM AND CIRCULANT GRAPH.

SOME PARAMETERS ON EQUITABLE COLORING OF PRISM AND CIRCULANT GRAPH. SOME PARAMETERS ON EQUITABLE COLORING OF PRISM AND CIRCULANT GRAPH. K VASUDEVAN, K. SWATHY AND K. MANIKANDAN 1 Dpartmnt of Mathmatics, Prsidncy Collg, Chnnai-05, India. E-Mail:vasu k dvan@yahoo.com. 2,

More information

COMPUTER GENERATED HOLOGRAMS Optical Sciences 627 W.J. Dallas (Monday, April 04, 2005, 8:35 AM) PART I: CHAPTER TWO COMB MATH.

COMPUTER GENERATED HOLOGRAMS Optical Sciences 627 W.J. Dallas (Monday, April 04, 2005, 8:35 AM) PART I: CHAPTER TWO COMB MATH. C:\Dallas\0_Courss\03A_OpSci_67\0 Cgh_Book\0_athmaticalPrliminaris\0_0 Combath.doc of 8 COPUTER GENERATED HOLOGRAS Optical Scincs 67 W.J. Dallas (onday, April 04, 005, 8:35 A) PART I: CHAPTER TWO COB ATH

More information

Estimation of apparent fraction defective: A mathematical approach

Estimation of apparent fraction defective: A mathematical approach Availabl onlin at www.plagiarsarchlibrary.com Plagia Rsarch Library Advancs in Applid Scinc Rsarch, 011, (): 84-89 ISSN: 0976-8610 CODEN (USA): AASRFC Estimation of apparnt fraction dfctiv: A mathmatical

More information

Square of Hamilton cycle in a random graph

Square of Hamilton cycle in a random graph Squar of Hamilton cycl in a random graph Andrzj Dudk Alan Friz Jun 28, 2016 Abstract W show that p = n is a sharp thrshold for th random graph G n,p to contain th squar of a Hamilton cycl. This improvs

More information

Fourier Transforms and the Wave Equation. Key Mathematics: More Fourier transform theory, especially as applied to solving the wave equation.

Fourier Transforms and the Wave Equation. Key Mathematics: More Fourier transform theory, especially as applied to solving the wave equation. Lur 7 Fourir Transforms and th Wav Euation Ovrviw and Motivation: W first discuss a fw faturs of th Fourir transform (FT), and thn w solv th initial-valu problm for th wav uation using th Fourir transform

More information

The Matrix Exponential

The Matrix Exponential Th Matrix Exponntial (with xrciss) by Dan Klain Vrsion 28928 Corrctions and commnts ar wlcom Th Matrix Exponntial For ach n n complx matrix A, dfin th xponntial of A to b th matrix () A A k I + A + k!

More information

Principles of Humidity Dalton s law

Principles of Humidity Dalton s law Principls of Humidity Dalton s law Air is a mixtur of diffrnt gass. Th main gas componnts ar: Gas componnt volum [%] wight [%] Nitrogn N 2 78,03 75,47 Oxygn O 2 20,99 23,20 Argon Ar 0,93 1,28 Carbon dioxid

More information

A Propagating Wave Packet Group Velocity Dispersion

A Propagating Wave Packet Group Velocity Dispersion Lctur 8 Phys 375 A Propagating Wav Packt Group Vlocity Disprsion Ovrviw and Motivation: In th last lctur w lookd at a localizd solution t) to th 1D fr-particl Schrödingr quation (SE) that corrsponds to

More information

CS 361 Meeting 12 10/3/18

CS 361 Meeting 12 10/3/18 CS 36 Mting 2 /3/8 Announcmnts. Homwork 4 is du Friday. If Friday is Mountain Day, homwork should b turnd in at my offic or th dpartmnt offic bfor 4. 2. Homwork 5 will b availabl ovr th wknd. 3. Our midtrm

More information

Derangements and Applications

Derangements and Applications 2 3 47 6 23 Journal of Intgr Squncs, Vol. 6 (2003), Articl 03..2 Drangmnts and Applications Mhdi Hassani Dpartmnt of Mathmatics Institut for Advancd Studis in Basic Scincs Zanjan, Iran mhassani@iasbs.ac.ir

More information

22/ Breakdown of the Born-Oppenheimer approximation. Selection rules for rotational-vibrational transitions. P, R branches.

22/ Breakdown of the Born-Oppenheimer approximation. Selection rules for rotational-vibrational transitions. P, R branches. Subjct Chmistry Papr No and Titl Modul No and Titl Modul Tag 8/ Physical Spctroscopy / Brakdown of th Born-Oppnhimr approximation. Slction ruls for rotational-vibrational transitions. P, R branchs. CHE_P8_M

More information

arxiv: v1 [cs.cg] 25 Jun 2012

arxiv: v1 [cs.cg] 25 Jun 2012 A Nw Uppr Bound for th VC-Dimnsion of Visibility Rgions arxi:1206.56891 [cs.cg] 25 Jun 2012 Alxandr Gilbrs Institut of Computr Scinc Unirsity of Bonn 53113 Bonn, Grmany gilbrs@cs.uni-bonn.d May 2, 2014

More information

SECTION where P (cos θ, sin θ) and Q(cos θ, sin θ) are polynomials in cos θ and sin θ, provided Q is never equal to zero.

SECTION where P (cos θ, sin θ) and Q(cos θ, sin θ) are polynomials in cos θ and sin θ, provided Q is never equal to zero. SETION 6. 57 6. Evaluation of Dfinit Intgrals Exampl 6.6 W hav usd dfinit intgrals to valuat contour intgrals. It may com as a surpris to larn that contour intgrals and rsidus can b usd to valuat crtain

More information

Deift/Zhou Steepest descent, Part I

Deift/Zhou Steepest descent, Part I Lctur 9 Dift/Zhou Stpst dscnt, Part I W now focus on th cas of orthogonal polynomials for th wight w(x) = NV (x), V (x) = t x2 2 + x4 4. Sinc th wight dpnds on th paramtr N N w will writ π n,n, a n,n,

More information

COUNTING TAMELY RAMIFIED EXTENSIONS OF LOCAL FIELDS UP TO ISOMORPHISM

COUNTING TAMELY RAMIFIED EXTENSIONS OF LOCAL FIELDS UP TO ISOMORPHISM COUNTING TAMELY RAMIFIED EXTENSIONS OF LOCAL FIELDS UP TO ISOMORPHISM Jim Brown Dpartmnt of Mathmatical Scincs, Clmson Univrsity, Clmson, SC 9634, USA jimlb@g.clmson.du Robrt Cass Dpartmnt of Mathmatics,

More information

Dealing with quantitative data and problem solving life is a story problem! Attacking Quantitative Problems

Dealing with quantitative data and problem solving life is a story problem! Attacking Quantitative Problems Daling with quantitati data and problm soling lif is a story problm! A larg portion of scinc inols quantitati data that has both alu and units. Units can sa your butt! Nd handl on mtric prfixs Dimnsional

More information

Addition of angular momentum

Addition of angular momentum Addition of angular momntum April, 0 Oftn w nd to combin diffrnt sourcs of angular momntum to charactriz th total angular momntum of a systm, or to divid th total angular momntum into parts to valuat th

More information

Roadmap. XML Indexing. DataGuide example. DataGuides. Strong DataGuides. Multiple DataGuides for same data. CPS Topics in Database Systems

Roadmap. XML Indexing. DataGuide example. DataGuides. Strong DataGuides. Multiple DataGuides for same data. CPS Topics in Database Systems Roadmap XML Indxing CPS 296.1 Topics in Databas Systms Indx fabric Coopr t al. A Fast Indx for Smistructurd Data. VLDB, 2001 DataGuid Goldman and Widom. DataGuids: Enabling Qury Formulation and Optimization

More information

Symmetric centrosymmetric matrix vector multiplication

Symmetric centrosymmetric matrix vector multiplication Linar Algbra and its Applications 320 (2000) 193 198 www.lsvir.com/locat/laa Symmtric cntrosymmtric matrix vctor multiplication A. Mlman 1 Dpartmnt of Mathmatics, Univrsity of San Francisco, San Francisco,

More information

On the irreducibility of some polynomials in two variables

On the irreducibility of some polynomials in two variables ACTA ARITHMETICA LXXXII.3 (1997) On th irrducibility of som polynomials in two variabls by B. Brindza and Á. Pintér (Dbrcn) To th mmory of Paul Erdős Lt f(x) and g(y ) b polynomials with intgral cofficints

More information

International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July ISSN

International Journal of Scientific & Engineering Research, Volume 6, Issue 7, July ISSN Intrnational Journal of Scintific & Enginring Rsarch, Volum 6, Issu 7, July-25 64 ISSN 2229-558 HARATERISTIS OF EDGE UTSET MATRIX OF PETERSON GRAPH WITH ALGEBRAI GRAPH THEORY Dr. G. Nirmala M. Murugan

More information

That is, we start with a general matrix: And end with a simpler matrix:

That is, we start with a general matrix: And end with a simpler matrix: DIAGON ALIZATION OF THE STR ESS TEN SOR INTRO DUCTIO N By th us of Cauchy s thorm w ar abl to rduc th numbr of strss componnts in th strss tnsor to only nin valus. An additional simplification of th strss

More information

A Prey-Predator Model with an Alternative Food for the Predator, Harvesting of Both the Species and with A Gestation Period for Interaction

A Prey-Predator Model with an Alternative Food for the Predator, Harvesting of Both the Species and with A Gestation Period for Interaction Int. J. Opn Problms Compt. Math., Vol., o., Jun 008 A Pry-Prdator Modl with an Altrnativ Food for th Prdator, Harvsting of Both th Spcis and with A Gstation Priod for Intraction K. L. arayan and. CH. P.

More information

Brief Introduction to Statistical Mechanics

Brief Introduction to Statistical Mechanics Brif Introduction to Statistical Mchanics. Purpos: Ths nots ar intndd to provid a vry quick introduction to Statistical Mchanics. Th fild is of cours far mor vast than could b containd in ths fw pags.

More information

Limiting value of higher Mahler measure

Limiting value of higher Mahler measure Limiting valu of highr Mahlr masur Arunabha Biswas a, Chris Monico a, a Dpartmnt of Mathmatics & Statistics, Txas Tch Univrsity, Lubbock, TX 7949, USA Abstract W considr th k-highr Mahlr masur m k P )

More information

SCHUR S THEOREM REU SUMMER 2005

SCHUR S THEOREM REU SUMMER 2005 SCHUR S THEOREM REU SUMMER 2005 1. Combinatorial aroach Prhas th first rsult in th subjct blongs to I. Schur and dats back to 1916. On of his motivation was to study th local vrsion of th famous quation

More information

PROOF OF FIRST STANDARD FORM OF NONELEMENTARY FUNCTIONS

PROOF OF FIRST STANDARD FORM OF NONELEMENTARY FUNCTIONS Intrnational Journal Of Advanc Rsarch In Scinc And Enginring http://www.ijars.com IJARSE, Vol. No., Issu No., Fbruary, 013 ISSN-319-8354(E) PROOF OF FIRST STANDARD FORM OF NONELEMENTARY FUNCTIONS 1 Dharmndra

More information

Ch. 24 Molecular Reaction Dynamics 1. Collision Theory

Ch. 24 Molecular Reaction Dynamics 1. Collision Theory Ch. 4 Molcular Raction Dynamics 1. Collision Thory Lctur 16. Diffusion-Controlld Raction 3. Th Matrial Balanc Equation 4. Transition Stat Thory: Th Eyring Equation 5. Transition Stat Thory: Thrmodynamic

More information

5.80 Small-Molecule Spectroscopy and Dynamics

5.80 Small-Molecule Spectroscopy and Dynamics MIT OpnCoursWar http://ocw.mit.du 5.80 Small-Molcul Spctroscopy and Dynamics Fall 008 For information about citing ths matrials or our Trms of Us, visit: http://ocw.mit.du/trms. Lctur # 3 Supplmnt Contnts

More information

From Elimination to Belief Propagation

From Elimination to Belief Propagation School of omputr Scinc Th lif Propagation (Sum-Product lgorithm Probabilistic Graphical Modls (10-708 Lctur 5, Sp 31, 2007 Rcptor Kinas Rcptor Kinas Kinas X 5 ric Xing Gn G T X 6 X 7 Gn H X 8 Rading: J-hap

More information

Computing and Communications -- Network Coding

Computing and Communications -- Network Coding 89 90 98 00 Computing and Communications -- Ntwork Coding Dr. Zhiyong Chn Institut of Wirlss Communications Tchnology Shanghai Jiao Tong Univrsity China Lctur 5- Nov. 05 0 Classical Information Thory Sourc

More information

Calculus concepts derivatives

Calculus concepts derivatives All rasonabl fforts hav bn mad to mak sur th nots ar accurat. Th author cannot b hld rsponsibl for any damags arising from th us of ths nots in any fashion. Calculus concpts drivativs Concpts involving

More information

3 Finite Element Parametric Geometry

3 Finite Element Parametric Geometry 3 Finit Elmnt Paramtric Gomtry 3. Introduction Th intgral of a matrix is th matrix containing th intgral of ach and vry on of its original componnts. Practical finit lmnt analysis rquirs intgrating matrics,

More information

Approximating the Two-Level Facility Location Problem Via a Quasi-Greedy Approach. Jiawei Zhang. October 03, 2003

Approximating the Two-Level Facility Location Problem Via a Quasi-Greedy Approach. Jiawei Zhang. October 03, 2003 Approximating th Two-Lvl Facility Location Problm Via a Quasi-Grdy Approach Jiawi Zhang Octobr 03, 2003 Abstract W propos a quasi-grdy algorithm for approximating th classical uncapacitatd 2-lvl facility

More information

Addition of angular momentum

Addition of angular momentum Addition of angular momntum April, 07 Oftn w nd to combin diffrnt sourcs of angular momntum to charactriz th total angular momntum of a systm, or to divid th total angular momntum into parts to valuat

More information

Elements of Statistical Thermodynamics

Elements of Statistical Thermodynamics 24 Elmnts of Statistical Thrmodynamics Statistical thrmodynamics is a branch of knowldg that has its own postulats and tchniqus. W do not attmpt to giv hr vn an introduction to th fild. In this chaptr,

More information

4. Money cannot be neutral in the short-run the neutrality of money is exclusively a medium run phenomenon.

4. Money cannot be neutral in the short-run the neutrality of money is exclusively a medium run phenomenon. PART I TRUE/FALSE/UNCERTAIN (5 points ach) 1. Lik xpansionary montary policy, xpansionary fiscal policy rturns output in th mdium run to its natural lvl, and incrass prics. Thrfor, fiscal policy is also

More information

EXST Regression Techniques Page 1

EXST Regression Techniques Page 1 EXST704 - Rgrssion Tchniqus Pag 1 Masurmnt rrors in X W hav assumd that all variation is in Y. Masurmnt rror in this variabl will not ffct th rsults, as long as thy ar uncorrlatd and unbiasd, sinc thy

More information

Rational Approximation for the one-dimensional Bratu Equation

Rational Approximation for the one-dimensional Bratu Equation Intrnational Journal of Enginring & Tchnology IJET-IJES Vol:3 o:05 5 Rational Approximation for th on-dimnsional Bratu Equation Moustafa Aly Soliman Chmical Enginring Dpartmnt, Th British Univrsity in

More information

u r du = ur+1 r + 1 du = ln u + C u sin u du = cos u + C cos u du = sin u + C sec u tan u du = sec u + C e u du = e u + C

u r du = ur+1 r + 1 du = ln u + C u sin u du = cos u + C cos u du = sin u + C sec u tan u du = sec u + C e u du = e u + C Tchniqus of Intgration c Donald Kridr and Dwight Lahr In this sction w ar going to introduc th first approachs to valuating an indfinit intgral whos intgrand dos not hav an immdiat antidrivativ. W bgin

More information

u x v x dx u x v x v x u x dx d u x v x u x v x dx u x v x dx Integration by Parts Formula

u x v x dx u x v x v x u x dx d u x v x u x v x dx u x v x dx Integration by Parts Formula 7. Intgration by Parts Each drivativ formula givs ris to a corrsponding intgral formula, as w v sn many tims. Th drivativ product rul yilds a vry usful intgration tchniqu calld intgration by parts. Starting

More information

ECE602 Exam 1 April 5, You must show ALL of your work for full credit.

ECE602 Exam 1 April 5, You must show ALL of your work for full credit. ECE62 Exam April 5, 27 Nam: Solution Scor: / This xam is closd-book. You must show ALL of your work for full crdit. Plas rad th qustions carfully. Plas chck your answrs carfully. Calculators may NOT b

More information

Where k is either given or determined from the data and c is an arbitrary constant.

Where k is either given or determined from the data and c is an arbitrary constant. Exponntial growth and dcay applications W wish to solv an quation that has a drivativ. dy ky k > dx This quation says that th rat of chang of th function is proportional to th function. Th solution is

More information

Some remarks on Kurepa s left factorial

Some remarks on Kurepa s left factorial Som rmarks on Kurpa s lft factorial arxiv:math/0410477v1 [math.nt] 21 Oct 2004 Brnd C. Kllnr Abstract W stablish a connction btwn th subfactorial function S(n) and th lft factorial function of Kurpa K(n).

More information

Problem Set 6 Solutions

Problem Set 6 Solutions 6.04/18.06J Mathmatics for Computr Scinc March 15, 005 Srini Dvadas and Eric Lhman Problm St 6 Solutions Du: Monday, March 8 at 9 PM in Room 3-044 Problm 1. Sammy th Shark is a financial srvic providr

More information

Equidistribution and Weyl s criterion

Equidistribution and Weyl s criterion Euidistribution and Wyl s critrion by Brad Hannigan-Daly W introduc th ida of a sunc of numbrs bing uidistributd (mod ), and w stat and prov a thorm of Hrmann Wyl which charactrizs such suncs. W also discuss

More information

ON RIGHT(LEFT) DUO PO-SEMIGROUPS. S. K. Lee and K. Y. Park

ON RIGHT(LEFT) DUO PO-SEMIGROUPS. S. K. Lee and K. Y. Park Kangwon-Kyungki Math. Jour. 11 (2003), No. 2, pp. 147 153 ON RIGHT(LEFT) DUO PO-SEMIGROUPS S. K. L and K. Y. Park Abstract. W invstigat som proprtis on right(rsp. lft) duo po-smigroups. 1. Introduction

More information

Background: We have discussed the PIB, HO, and the energy of the RR model. In this chapter, the H-atom, and atomic orbitals.

Background: We have discussed the PIB, HO, and the energy of the RR model. In this chapter, the H-atom, and atomic orbitals. Chaptr 7 Th Hydrogn Atom Background: W hav discussd th PIB HO and th nrgy of th RR modl. In this chaptr th H-atom and atomic orbitals. * A singl particl moving undr a cntral forc adoptd from Scott Kirby

More information

ON THE DISTRIBUTION OF THE ELLIPTIC SUBSET SUM GENERATOR OF PSEUDORANDOM NUMBERS

ON THE DISTRIBUTION OF THE ELLIPTIC SUBSET SUM GENERATOR OF PSEUDORANDOM NUMBERS INTEGERS: ELECTRONIC JOURNAL OF COMBINATORIAL NUMBER THEORY 8 2008, #A3 ON THE DISTRIBUTION OF THE ELLIPTIC SUBSET SUM GENERATOR OF PSEUDORANDOM NUMBERS Edwin D. El-Mahassni Dpartmnt of Computing, Macquari

More information

Abstract Interpretation: concrete and abstract semantics

Abstract Interpretation: concrete and abstract semantics Abstract Intrprtation: concrt and abstract smantics Concrt smantics W considr a vry tiny languag that manags arithmtic oprations on intgrs valus. Th (concrt) smantics of th languags cab b dfind by th funzcion

More information

Outerplanar graphs and Delaunay triangulations

Outerplanar graphs and Delaunay triangulations CCCG 011, Toronto ON, August 10 1, 011 Outrplanar graphs and Dlaunay triangulations Ashraful Alam Igor Rivin Ilana Strinu Abstract Ovr 0 yars ago, Dillncourt [1] showd that all outrplanar graphs can b

More information

ME 321 Kinematics and Dynamics of Machines S. Lambert Winter 2002

ME 321 Kinematics and Dynamics of Machines S. Lambert Winter 2002 3.4 Forc Analysis of Linkas An undrstandin of forc analysis of linkas is rquird to: Dtrmin th raction forcs on pins, tc. as a consqunc of a spcifid motion (don t undrstimat th sinificanc of dynamic or

More information

What are those βs anyway? Understanding Design Matrix & Odds ratios

What are those βs anyway? Understanding Design Matrix & Odds ratios Ral paramtr stimat WILD 750 - Wildlif Population Analysis of 6 What ar thos βs anyway? Undrsting Dsign Matrix & Odds ratios Rfrncs Hosmr D.W.. Lmshow. 000. Applid logistic rgrssion. John Wily & ons Inc.

More information

Function Spaces. a x 3. (Letting x = 1 =)) a(0) + b + c (1) = 0. Row reducing the matrix. b 1. e 4 3. e 9. >: (x = 1 =)) a(0) + b + c (1) = 0

Function Spaces. a x 3. (Letting x = 1 =)) a(0) + b + c (1) = 0. Row reducing the matrix. b 1. e 4 3. e 9. >: (x = 1 =)) a(0) + b + c (1) = 0 unction Spacs Prrquisit: Sction 4.7, Coordinatization n this sction, w apply th tchniqus of Chaptr 4 to vctor spacs whos lmnts ar functions. Th vctor spacs P n and P ar familiar xampls of such spacs. Othr

More information

Difference -Analytical Method of The One-Dimensional Convection-Diffusion Equation

Difference -Analytical Method of The One-Dimensional Convection-Diffusion Equation Diffrnc -Analytical Mthod of Th On-Dimnsional Convction-Diffusion Equation Dalabav Umurdin Dpartmnt mathmatic modlling, Univrsity of orld Economy and Diplomacy, Uzbistan Abstract. An analytical diffrncing

More information

The second condition says that a node α of the tree has exactly n children if the arity of its label is n.

The second condition says that a node α of the tree has exactly n children if the arity of its label is n. CS 6110 S14 Hanout 2 Proof of Conflunc 27 January 2014 In this supplmntary lctur w prov that th λ-calculus is conflunt. This is rsult is u to lonzo Church (1903 1995) an J. arkly Rossr (1907 1989) an is

More information

Two Products Manufacturer s Production Decisions with Carbon Constraint

Two Products Manufacturer s Production Decisions with Carbon Constraint Managmnt Scinc and Enginring Vol 7 No 3 pp 3-34 DOI:3968/jms9335X374 ISSN 93-34 [Print] ISSN 93-35X [Onlin] wwwcscanadant wwwcscanadaorg Two Products Manufacturr s Production Dcisions with Carbon Constraint

More information

GEOMETRICAL PHENOMENA IN THE PHYSICS OF SUBATOMIC PARTICLES. Eduard N. Klenov* Rostov-on-Don, Russia

GEOMETRICAL PHENOMENA IN THE PHYSICS OF SUBATOMIC PARTICLES. Eduard N. Klenov* Rostov-on-Don, Russia GEOMETRICAL PHENOMENA IN THE PHYSICS OF SUBATOMIC PARTICLES Eduard N. Klnov* Rostov-on-Don, Russia Th articl considrs phnomnal gomtry figurs bing th carrirs of valu spctra for th pairs of th rmaining additiv

More information

Tight Bounds on the Round Complexity of the Distributed Maximum Coverage Problem

Tight Bounds on the Round Complexity of the Distributed Maximum Coverage Problem Tight Bounds on th Round Complxity of th Distributd Maximum Covrag Problm Sphr Assadi Sanjv Khanna Abstract W study th maximum k-st covrag problm in th following distributd stting. A collction of input

More information

arxiv: v3 [cs.ds] 14 Jul 2016

arxiv: v3 [cs.ds] 14 Jul 2016 Ngativ-Wight Shortst Paths and Unit Capacity Minimum Cost Flow in Õ m 0/7 log W Tim Michal B. Cohn Alksandr Mądry Piotr Sankowski Adrian Vladu MIT MIT Univrsity of Warsaw MIT micohn@mit.du madry@mit.du

More information

The van der Waals interaction 1 D. E. Soper 2 University of Oregon 20 April 2012

The van der Waals interaction 1 D. E. Soper 2 University of Oregon 20 April 2012 Th van dr Waals intraction D. E. Sopr 2 Univrsity of Orgon 20 pril 202 Th van dr Waals intraction is discussd in Chaptr 5 of J. J. Sakurai, Modrn Quantum Mchanics. Hr I tak a look at it in a littl mor

More information

General Notes About 2007 AP Physics Scoring Guidelines

General Notes About 2007 AP Physics Scoring Guidelines AP PHYSICS C: ELECTRICITY AND MAGNETISM 2007 SCORING GUIDELINES Gnral Nots About 2007 AP Physics Scoring Guidlins 1. Th solutions contain th most common mthod of solving th fr-rspons qustions and th allocation

More information

surface of a dielectric-metal interface. It is commonly used today for discovering the ways in

surface of a dielectric-metal interface. It is commonly used today for discovering the ways in Surfac plasmon rsonanc is snsitiv mchanism for obsrving slight changs nar th surfac of a dilctric-mtal intrfac. It is commonl usd toda for discovring th was in which protins intract with thir nvironmnt,

More information

Final Exam Solutions

Final Exam Solutions CS 2 Advancd Data Structurs and Algorithms Final Exam Solutions Jonathan Turnr /8/20. (0 points) Suppos that r is a root of som tr in a Fionacci hap. Assum that just for a dltmin opration, r has no childrn

More information

Cramér-Rao Inequality: Let f(x; θ) be a probability density function with continuous parameter

Cramér-Rao Inequality: Let f(x; θ) be a probability density function with continuous parameter WHEN THE CRAMÉR-RAO INEQUALITY PROVIDES NO INFORMATION STEVEN J. MILLER Abstract. W invstigat a on-paramtr family of probability dnsitis (rlatd to th Parto distribution, which dscribs many natural phnomna)

More information

Introduction to the quantum theory of matter and Schrödinger s equation

Introduction to the quantum theory of matter and Schrödinger s equation Introduction to th quantum thory of mattr and Schrödingr s quation Th quantum thory of mattr assums that mattr has two naturs: a particl natur and a wa natur. Th particl natur is dscribd by classical physics

More information

Answer Homework 5 PHA5127 Fall 1999 Jeff Stark

Answer Homework 5 PHA5127 Fall 1999 Jeff Stark Answr omwork 5 PA527 Fall 999 Jff Stark A patint is bing tratd with Drug X in a clinical stting. Upon admiion, an IV bolus dos of 000mg was givn which yildd an initial concntration of 5.56 µg/ml. A fw

More information