A Comparative Study of Constraint-Handling Techniques in Evolutionary Constrained Multiobjective Optimization

Size: px
Start display at page:

Download "A Comparative Study of Constraint-Handling Techniques in Evolutionary Constrained Multiobjective Optimization"

Transcription

1 A omparatve Stdy of onstrant-handlng Technqes n Evoltonary onstraned Mltobectve Optmzaton Ja-Peng L, Yong Wang, Member, IEEE, Shengxang Yang, Senor Member, IEEE, and Zxng a, Senor Member, IEEE Abstract Solvng constraned mltobectve optmzaton problems s one of the most challengng areas n the evoltonary comptaton research commnty. To solve a constraned mltobectve optmzaton problem, an algorthm shold tackle the obectve fnctons and the constrants smltaneosly. As a reslt, many constrant-handlng technqes have been proposed. However, most of the exstng constrant-handlng technqes are developed to solve test nstances (e.g., TPs) wth low dmenson and large feasble regon. On the other hand, expermental comparsons on dfferent constrant-handlng technqes reman scarce. In vew of these two sses, n ths paper we frst constrct 8 test nstances, each of whch exhbts dfferent propertes. Afterward, we choose three representatve constrant-handlng technqes and combne them wth nondomnated sortng genetc algorthm II to stdy the performance dfference on varos condtons. By the expermental stdes, we pont ot the advantages and dsadvantages of dfferent constrant-handlng technqes. Keywords onstrant-handlng technqes, constraned mltobectve optmzaton problems, evoltonary algorthms, test nstances. I. INTRODUTION Nmeros real-world applcatons nvolve mltple obectves and varos constrants, whch can be modeled as constraned mltobectve optmzaton problems (MOPs). Wthot loss of generalty, a MOP can be formlated as: Mnmze f ) ( f), f2 ),, fm )), x,, xn ) S Sbect to: g ) 0,,..., p h ) 0, p,..., q where x s the decson vector, L x U ( {,, n}) s the th decson varable, L and U are the lower and pper bonds of x, S s the decson space, n and m are the nmber of decson varables and obectve fnctons, f ) ( {,, m}) s the th obectve fncton, g ) s the th neqalty constrant, h ) s the ( p)th eqalty constrant, p s the nmber of neqalty constrants, and ( q p) s the nmber of eqalty constrants. J.-P. L and Z. a are wth the School of Informaton Scence and Engneerng, entral Soth Unversty, hangsha 40083, hna. (e-mal: lpcs@cs.ed.cn; zxca@cs.ed.cn) Y. Wang (orrespondng Athor) s wth the School of Informaton Scence and Engneerng, entral Soth Unversty, hangsha 40083, hna, and also wth the School of ompter Scence and Informatcs, De Montfort Unversty, Lecester LE 9BH, UK. (e-mal: ywang@cs.ed.cn) S. Yang s wth the School of ompter Scence and Informatcs, De Montfort Unversty, Lecester LE 9BH, UK. (e-mal: syang@dm.ac.k) The constrant volaton of ndvdal x on the th constrant s calclated as follows: (0, g )), p ) () (0, h ) ), p q where s a predefned tolerance vale to relax the eqalty constrans to a certan extent. The feasble regon of a MOP s a sbspace of the decson space S, and can be defned as { x S ) 0,,, q}. Evoltonary algorthms (EAs) are poplaton based optmzaton approaches nspred by natre []. Drng the last two decades, mch effort has been made to develop and nderstand EAs for dealng wth mltobectve optmzaton problems [2] [3]. onstrant-handlng for sngle-obectve optmzaton problems has also been extensvely researched [4] [5]. However, few attempts have been made to nvestgate evoltonary constraned mltobectve optmzaton [6], whch s challengng and of practcal nterest. For MOPs, we cannot fnd a sngle solton to optmze all the obectves at the same tme snce the obectves are always n conflct wth each other. Therefore, when sng EAs to deal wth MOPs, we have to balance the obectves and fnd a set of optmal tradeoffs. In addton, t s necessary to note that EAs are nconstraned search methods that need addtonal mechansms to deal wth constrants when solvng MOPs. De to the above attrbtes, the solton of MOPs s very dffclt and the research of MOPs based on EAs s stll n ts nfant stage. When solvng MOPs by EAs, the correspondng methods are called constraned mltobectve EAs (MOEAs). Next, we wll brefly ntrodce some representatve MOEAs. Deb et al. [7] extended the feasblty rle n [4] and proposed a constraned-domnaton prncple (DP) to solve MOPs. Based on the work n [7], Oyama et al. [8] proposed a new constrant-handlng technqe, whch ntrodces the dea of nondomnance and nchng concepts n the obectve space nto the constrant space. hafekar et al. [9] ntrodced a steady state GA to solve MOPs. In [9], two constraned mltobectve optmzaton technqes are appled. One technqe s to rn several sngle-obectve GAs concrrently and each GA s to optmze one obectve. The other technqe s to se a sngle-obectve GA to optmze mltple obectves n a seqental order. Yong [0] presented a blended rank mechansm n whch every solton s assgned two Pareto domnance ranks, and both two ranks are taken nto accont to gve the fnal rank of an ndvdal. Jmenez et al. [] combned the Pareto concept n mltobectve optmzaton

2 wth a novel dversty mechansm to sort the poplaton. Woldesenbet and Yen [6] proposed a self-adaptve penalty fncton to handle constrants based on the feasble rato n the crrent poplaton. Santana-Qntero et al. [2] presented a hybrd MOEA by approxmatng the Pareto front of a MOP wth a hybrd approach. In [2], rogh set theory s mplemented to mantan a good qalty of the approxmaton. By combnng DP [7] wth smlated annealng, Sngh et al. [3] proposed a non-greedy constrant-handlng technqe to solve MOPs. Jao et al. [4] proposed a MOEA based on [6], the man deas of whch are to make se of the nformaton of the nfeasble ndvdals and to constrct some modfed obectve fnctons to gde the evoltonary process. Jan and Khanm [5] proposed two modfed constrant-handlng technqes by combnng Tchebycheff aggregaton fncton wth the famos stochastc rankng [6] and DP [7] to solve MOPs. Very recently, an mmne optmzaton algorthm s proposed by Qan et al. [7] to solve MOPs. Takng both convergence and dversty nto accont, they dvded a poplaton nto two sbpoplatons and appled dfferent reprodcton operators to create the next poplaton. Recognzng that there s no sngle constrant-handlng technqe can otperform all others on dfferent knds of problems, Q and Sganthan [8] proposed an ensemble of constrant-handlng technqes. In [8], three constranthandlng technqes,.e., DP [7], self-adaptve penalty (SP) [6], and ε constraned method [9] are adopted. It s noteworthy that most of the above methods are desgned to solve the test nstances (called TPs) n [20], whch were proposed by Deb et al. ffteen years ago. De to the fact that TPs sally have a large proporton of the feasble regon, t s a very mportant topc to desgn MOPs wth more complex characterstcs. On the other hand, the crrent methods are sally developed to solve TPs wth low dmenson (sch as two dmensons), and there are very few stdes condcted on the performance comparson of dfferent constrant-handlng technqes n the commnty of evoltonary constraned mltobectve optmzaton. Motvated by the above consderatons, n ths paper we frstly constrct 8 test nstances wth dfferent characterstcs based on TPs. Afterward, we select three representatve constrant-handlng technqes (DP [7], SP [6], and adaptve tradeoff model (ATM) [2]). Then we stdy ther performance dfference on varos scenaros. The man contrbtons of ths paper can be smmarzed as follows: We desgn 8 new test nstances (called NTPs) based on TPs. ompared wth the orgnal TPs, NTPs ntrodced n ths paper exhbt more complex characterstcs. Specfcally, n NTPs dfferent test nstances have dfferent shapes of the Pareto front, dfferent dmensons of the search space, and dfferent sze of the feasble regon. Systematc experments have been condcted on the 8 test nstances to stdy the performance of the three representatve constrant-handlng technqes accordng to the shape of the Pareto front, the dmenson of search space, and the sze of the feasble regon. The rest of ths paper s organzed as follows. Secton II brefly descrbes the related defntons. Secton III ntrodces three representatve constrant-handlng technqes. Secton IV presents the constrcted test nstances. Secton V expermentally stdes the performance of the chosen constranthandlng technqes. Secton VI concldes ths paper. II. THE RELATED DEFINITIONS OF MOPS A MOP nclde mltple obectves. In mltobectve optmzaton, the comparson of ndvdals s sally based on Pareto domnance. Next, we wll ntrodce for related defntons. Defnton (Pareto Domnance): onsderng the m obectve fnctons f ) ( f), f2 ),, fm )) and two decson vectors x v and x, f {,..., m}, f v ) f ) and {,..., m}, f v ) f ), then x s sad to v Pareto domnate x, denoted as xv x. Defnton 2 (Pareto Optmal Solton): A solton x s called a Pareto optmal solton of a MOP f and only f, x x. x v v Defnton 3 (Pareto Optmal Set): The Pareto optmal set of a MOP can be defned as POS { x xv, x x v }. Defnton 4 (Pareto Front): The Pareto front of a MOP can be defned as PF { f ) x POS}. III. THREE REPRESENTATIVE ONSTRAINT-HANDLING TEHNIQUES FOR MOPS A. onstraned-domnaton Prncple (DP) DP [7] s a smple and effcent technqe to handle constrans, whch compares parwse ndvdals based on the followng rles: When two feasble soltons are compared, the one Pareto domnatng the other s better. When a feasble solton s compared wth an nfeasble solton, the feasble solton s better. When two nfeasble soltons are compared, the one wth smaller degree of constrant volaton s better. B. Self-adaptve Penalty (SP) Another way to solve MOPs s to penalze the nfeasble ndvdal wth penalty fncton, n whch the penalty term added to the obectve fncton s based on the degree of constrant volaton of the nfeasble ndvdal. Among all penalty fncton based methods, the self-adaptve penalty (SP) proposed n [6] s a representatve one. It has two man components: the dstance vale and the penalty fncton. Frstly, the obectve fnctons and the degree of constrant

3 volaton of each ndvdal x n the poplaton are normalzed as follows: mn f ) f f ), {,, m} (2) mn f f q ) ) (3) q where mn f s the mnmm vale of the th obectve fncton, f s the mm vale of the th obectve fncton, and s the mm volaton of the th constrant. Let r f denote the feasblty rato of the crrent poplaton, whch can be calclated as follows: the nmber of feasble ndvdals rf (4) the poplaton sze Afterward, the th ( {,, m}) dstance vale d ) and th ( {,, m}) penalty vale p ) of x can be expressed as follows: ), f rf 0 d ) 2 2 f ( ) ( ), f rf 0 x x (5) p ) ( r ) X ) r Y ) (6) f f where 0, f rf 0 X ) (7) ), f rf 0 0, f x s feasble Y ) f (8) ( ), f x s nfeasble x Fnally, the ftness of x n the th obectve fncton dmenson s the sm of d ) and p ) : F ) d ) p ), {,, m} (9) Then the poplaton s sorted based on the m ftness fnctons F, F2,, Fm va the nondomnated sortng [7].. Adaptve Tradeoff Model (ATM) ATM ntrodces an mportant dea,.e., dvdng the evoltonary process nto three statons by the feasblty proporton of the crrent poplaton. In dfferent statons, ATM adopts dfferent technqes to cope wth the constrants. ) The nfeasble staton: There s no feasble solton n the crrent poplaton. ATM converts a MOP wth m obectves and q constrants nto a nconstraned MOP wth (m+) obectves, by consderng the constrant volaton as an addtonal obectve. Then the nondomnated sortng [7] s appled, and half of the ndvdals wth less constrant volatons n the frst layer are chosen and removed from the poplaton. Afterward, the remanng ndvdals n the poplaton are mplemented the same process ntl a desrable nmber of ndvdals s obtaned. 2) The sem-feasble staton: There exst both nfeasble and feasble soltons n the crrent poplaton. Assme that Z s the crrent poplaton, and then Z s dvded nto the feasble grop and the nfeasble grop based on eqaton (3): Z { x Z ) 0} (0) Z2 { x Z ) 0} () Frstly, ATM ses the followng transformed obectve fncton to penalze the nfeasble ndvdals: ( ), s feasble ' f x f x f ) { * fb _ ( )* fw _, f )}, f x s nfeasble (2) where denotes the feasblty proporton of the last poplaton, mn ( f ) b _ f x, and ( f ). w xz _ f x xz Then, ATM normalzes the transformed obectve fncton and the constrant volaton: ' ' f ) mn f ) x Z f ), {,, } ' ' m (3) f ) mn f ) xz xz 0, f x Z ) ( ) mn ( ) x x xz (4) 2, f x Z 2 ) mn ) xz2 xz2 The fnal ftness of x n the th obectve fncton dmenson s the sm of f ) and ) : F ) f ) ), {,, m} (5) Lke SP, the m ftness fnctons F, F2,, Fm are sed to sort the poplaton by the nondomnated sortng [7]. 3) The feasble phase: In ths phase, all soltons n the poplaton are feasble. Ths, ATM drectly sorts the poplaton by makng se of the nondomnated sortng [7]. IV. ONSTRUTED TEST INSTANES The wdely sed test nstances n evoltonary constraned mltobectve optmzaton are TPs [20], whch can be dvded n two parts. The frst part s TP, n whch the nmber of constrants s changeable, bt wth the ncrease of constrants the shape of the Pareto front almost remans the same. The second part s TP2-TP7. The shapes of the Pareto front of these test nstances are tnable and controlled by sx parameters, bt these test nstances only have one constrant and the sze of the feasble regon s sally large. To stdy the performance of the three chosen constranthandlng technqes n Secton III on dfferent condtons, the above drawbacks of test nstances shold be addressed. In ths paper, a set of new constraned test nstances (called NTPs) s constrcted based on TP2-TP7. In NTPs, the followng new characterstcs have been added: In most of papers, g ) n the second obectve fncton of TP2-TP7 s often set to (+x 2 ). To ncrease the dffclty, n ths paper g ) has been set to the Ronsenbrock fncton [22]: n g) (00 x ) ( x ) ) (6) 2

4 TABLE I Detaled nformaton of the 8 test nstances, where Type I denotes that the Pareto front s dscontnos (see Fg. ), Type II denotes that the Pareto front conssts of both dscontnos and contnos parts (see Fg. 2), and Type III denotes that the Pareto front s contnos (see Fg. 3). Test nstance Parameter vale onstrant Shape of the Pareto front Propertes Dmenson of the decson vector Sze of the feasble regon NTP- 0.2, a 0.2, b 0, c, d 0.5, e, z 0.5, z2 4 and 2 Type I 0D/30D Small (<0.%) NTP-2 0.2, a 0.75, b 0, c, d 0.5, e, z 0.5, z2 4 and 2 Type I 0D/30D Small (<0.%) NTP-3 0.2, a 2, b 0, c, d 6, e, z 0.5, z2 6 and 2 Type I 0D/30D Small (<0.%) NTP-4 0.2, a 0.2, b 0, c, d 0.5, e, z 0.5 Type I 0D/30D Large (>99.9%) NTP-5 0.2, a 0.75, b 0, c, d 0.5, e, z 0.5 Type I 0D/30D Large (>99.9%) NTP-6 0.2, a 2, b 0, c, d 6, e, z 0.5 Type I 0D/30D Large (>99.9%) NTP-7 0.2, a 0.2, b 0, c, d 0.5, e, z, z2 4 and 2 Type II 0D/30D Small (<0.%) NTP-8 0.2, a 0.75, b 0, c, d 0.5, e, z, z2 4 and 2 Type II 0D/30D Small (<0.%) NTP-9 0.2, a 2, b 0, c, d 6, e, z, z2 6 and 2 Type II 0D/30D Small (<0.%) NTP-0 0.2, a 0.2, b 0, c, d 0.5, e, z Type II 0D/30D Large (>99.9%) NTP- 0.2, a 0.75, b 0, c, d 0.5, e, z Type II 0D/30D Large (>99.9%) NTP-2 0.2, a 2, b 0, c, d 6, e, z Type II 0D/30D Large (>99.9%) NTP-3 0.2, a 0.2, b 0, c, d 0.5, e, z 2, z2 4 and 2 Type III 0D/30D Small (<0.%) NTP-4 0.2, a 0.75, b 0, c, d 0.5, e, z 2.5, z2 4 and 2 Type III 0D/30D Small (<0.%) NTP-5 0.2, a 2, b 0, c, d 6, e, z 4, z2 6 and 2 Type III 0D/30D Small (<0.%) NTP-6 0.2, a 0.2, b 0, c, d 0.5, e, z 2 Type III 0D/30D Large (>99.9%) NTP-7 0.2, a 0.75, b 0, c, d 0.5, e, z 2.5 Type III 0D/30D Large (>99.9%) NTP-8 0.2, a 2, b 0, c, d 6, e, z 4 Type III 0D/30D Large (>99.9%) Meanwhle, the decson space has been changed from [0,] n n to [0,5]. An addtonal constrant ( ) 2 ) has been added to adst the sze of the feasble regon. An addtonal parameter ( z ) has been added to the second obectve fncton to control the proporton of the contnos part of the Pareto front. f ) has been changed to f ) n the second obectve, wth the am of makng the nconstraned Pareto front a crve rather than a lne segment. The proposed test nstances can be descrbed as follows: Mnmze f) x f) Mnmze f2) g)( ) z g) sbect to ) cos( )( f2) e) sn( ) f) c d a sn( b (sn( )( f2) e) cos( ) f)) ) ) f ) 0.73 f ) z (7) where, a, b, c, d, and e are sed to control the topology of the Pareto front [20], z s sed to control the proporton of the contnos part of the Pareto front, and z 2 s sed to control the sze of the feasble regon. Based on the above formlaton, 8 test nstances have been desgned n ths paper to test the performance of the three representatve constrant-handlng technqes. The detals of these test nstances have been presented n Table I, and the obectve spaces of these test nstances have been presented n Fg., Fg. 2, and Fg. 3. A. Expermental Setp V. EXPERIMENTAL STUDY In ths secton, we choose the nondomnated sortng genetc algorthm II (NSGA-II) [7] as the mltobectve optmzaton framework, de to ts hgh robstness and ease of mplementaton. Then the three representatve constrant-handlng technqes (DP, SP, and ATM) ntrodced n Secton III are combned wth NSGA-II to solve the 8 test nstances proposed n Secton IV. For all the experments, the parameters were set as follows: the poplaton sze was 00, the crossover rate was 0.8, the mtaton rate was /n, and the mm nmber of generaton was 500. Besdes, the tornament selecton, smlated bnary crossover, and polynomal mtaton were adopted as the genetc operators n each smlaton. As presented n Table I, ths paper ntends to stdy how the followng three propertes of the 8 test nstances nflence the performance of a constrant-handlng technqe for MOPs. The shape of Pareto front The dmenson of decson vector The sze of feasble regon

5 Fg.. The obectve spaces of sx test nstances n type I, where the shade ndcates the feasble regon n the obectve space, the black crcle lne ndcates the nconstraned Pareto front, and the red dotted lne ndcates the constraned Pareto front. Fg. 2. The obectve spaces of sx test nstances n type II, where the shade ndcates the feasble regon n the obectve space, the black crcle lne ndcates the nconstraned Pareto front, and the red dotted lne ndcates the constraned Pareto front. Fg. 3. The obectve spaces of sx test nstances n type III, where the shade ndcates the feasble regon n the obectve space, the black crcle lne ndcates the nconstraned Pareto front, and the red dotted lne ndcates the constraned Pareto front. In order to stdy the effect of the second property, we tested two scenaros: n=0 and n=30. For each scenaro, 00 ndependent rns were mplemented for each test nstance. Accordng to or observaton, the performance of an algorthm sgnfcantly degenerates wth the ncrease of the dmenson of the decson vector. More mportantly,

6 TABLE II Test nstances nclded n each case: 0D-S ncldes the test nstances wth n=0 and small feasble regons, 30D-S ncldes the test nstances wth n=30 and small feasble regons, 0D-L ncldes the test nstances wth n=0 and large feasble regons, and 30D-L ncldes the test nstances wth n=30 and large feasble regons. Type Type I Type II Type III ase 0D-S 30D-S 0D-L 30D-L 0D-S 30D-S 0D-L 30D-L 0D-S 30D-S 0D-L 30D-L Test nstance NTP- NTP-4 NTP-7 NTP-0 NTP-3 NTP-6 NTP-2 NTP-5 NTP-8 NTP- NTP-4 NTP-7 NTP-3 NTP-6 NTP-9 NTP-2 NTP-5 NTP-8 Fg. 4. The expermental reslts of the for cases n type I Fg. 5. The expermental reslts of the for cases n type II Fg. 6. The expermental reslts of the for cases n type III sometmes an algorthm cannot fnd any feasble solton on the nne test nstances wth a small feasble regon when the evolton halts. The nverted generatonal dstance (IGD) [23] s a wdely sed ndcator to evalate the performance of an algorthm for mltobectve optmzaton. Note that f an algorthm cannot fnd any feasble solton, we cannot obtan the IGD vale. Ths, IGD mght not be stable to assess the performance of an algorthm for MOPs wth a small feasble regon. To ths end, we propose a revsed IGD ndcator, called IGD dstrbton ndcator (IGDD) n ths paper. The IGDD vale of an algorthm s compted as follows: Step ): alclate the IGD vale n each rn. As a reslt, we obtan 00 IGD vales n 00 rns. Step 2): Defne two ranges of the IGD vale: range (0, 0.25] and range2 (0.25, 0.5], compte the nmber (denoted as n ) of the IGD vale belongng to range, and compte the nmber (denoted as n 2 ) of the IGD vale belongng to range. 2 Note that n n2 00. Step 3): alclate the IGDD vale by the followng eqaton: IGDD= n ( ) n (8) 2 where s an coeffcent that s sed to adst the weght of n and n. Snce the IGD vale n 2 range s better than that n range, 2 n shold be pt more emphass. In ths paper, was set to 0.8. The IGDD ndcator can be sed to measre the convergence towards the Pareto front. The hgher the IGDD vale, the better the performance of an algorthm. B. omparatve Stdy For the sake of clarty, the comparatve stdy s based on the three dfferent types of shapes of the Pareto front ntrodced n Table I:

7 Fg fnal Pareto fronts obtaned by ATM, DP, and SP (from left to rght) on NTP-4 n type I, whch belongs to 30D-L. Fg fnal Pareto fronts obtaned by ATM, DP, and SP (from left to rght) on NTP- n type II, whch belongs to 0D-L. Fg fnal Pareto fronts obtaned by ATM, DP, and SP (from left to rght) on NTP-3 n type III, whch belongs to 30D-S. Type I denotes that the Pareto front s dscontnos (NTP--NTP-6). Type II denotes that the Pareto front conssts of both dscontnos part and contnos part (NTP-7- NTP-2). Type III denotes that the Pareto front s contnos (NTP-3-NTP-8). Based on the dmenson of the decson vector and the sze of the feasble regon, each type can be dvded nto for cases (0D-S, 30D-S, 0D-L, and 30D-L) as shown n Table II. The IGDD vales n the same case of each type are added together to represent the overall performance of a constrant-handlng technqe (see Fgs. 4-6). For example, the expermental reslt of ATM n 0D-S of Fg. 4 denotes the sm of the IGDD vales on NTP- wth 0D, NTP-2 wth 0D, and NTP-3 wth 0D. From the expermental reslts of Fgs. 4-6, we can gve the followng comments: Type I: As shown n Fg. 4, DP and SP perform smlarly n all the for cases. For the test nstances wth 0D, DP and SP performs mch better than ATM, regardless of the sze of feasble regon. However, ATM otperforms DP and SP on the test nstances wth 30D. Fg. 7 frther llstrates ths phenomenon. Ths fgre shows that the soltons obtaned by ATM are more close to the Pareto front and have a better dstrbton. ompared wth ATM, some soltons fond by DP are far away from the Pareto front, and the soltons obtaned by SP are not well dstrbted. Type II: Smlar to type I, n type II DP and SP srpass ATM on the test nstances wth low dmenson (0D). Fg. 8 provdes the performance comparson of ATM, DP, and SP on NTP- wth 0D. However, ATM and SP are slghtly better than DP on the test nstances wth 30D and small feasble regons. Moreover, ATM ranks the frst on the test nstances wth 30D and large feasble regons. Type III: In the case of 0D-S, DP s better than SP and ATM. In terms of 30D-S, SP ranks the frst, followed by ATM, whch s frther llstrated n Fg. 9. Wth regard to 0D-L and 30D-L, the three

8 constrant-handlng technqes show comparable performance. Overall, the ncrease of the nmber of decson varables poses a grand challenge to the three constrant-handlng technqes. In comparson wth the IGDD vales of the 0D test nstances, the IGDD vales of the 30D test nstances decrease sgnfcantly. VI. ONLUSION AND FUTURE WORK Accordng to the prelmnary expermental comparson n Secton V, the followng conclsons can be made accordng to the IGDD vales of the three constrant-handlng technqes: DP and SP show comparable performance on all test nstances, and they otperform ATM on test nstances wth 0D. ATM exhbts compettve performance on test nstances wth 30D. It s necessary to note that the above conclsons are obtaned by experments. In the ftre, we wll frther analyze the advantages and dsadvantages of the constrant-handlng technqes on dfferent knds of MOPs n prncple. On the other hand, we wll desgn new constrant-handlng technqes to tackle the 8 test nstances developed n ths paper. The Matlab sorce code can be downloaded from Y. Wang s homepage: AKNOWLEDGMENTS Ths work was spported n part by the Innovaton-drven Plan n entral Soth Unversty (No. 205XS02 and No. 205X007), n part by the Natonal Natral Scence Fondaton of hna nder Grant , n part by the EU Horzon 2020 Mare Sklodowska-re Indvdal Fellowshps (Proect ID: 66327), n part by the Engneerng and Physcal Scences Research oncl of UK nder Grant EP/K0030/, n part by the Hnan Provncal Natral Scence Fnd for Dstngshed Yong Scholars (Grant No. 206JJ08), and n part by the Program for New entry Excellent Talents n Unversty nder Grant NET REFERENES [] T. Bäck, D. B. Fogel, and Z. Mchalewcz, Handbook of Evoltonary omptaton. IOP Pblshng Ltd., 997. [2] K. Deb, Mlt-obectve Optmzaton Usng Evoltonary Algorthms. John Wley & Sons, 200. [3]. A. oello oello, D. A. Van Veldhzen, and G. B. Lamont, Evoltonary Algorthms for Solvng Mlt-Obectve Problems. New York: Klwer Academc, [4] K. Deb, An effcent constrant handlng method for genetc algorthms, ompter Methods n Appled Mechancs and Engneerng, vol. 86, no. 2-4, pp , [5] Z. a and Y. Wang, A mltobectve optmzaton based evoltonary algorthm for constraned optmzaton, IEEE Transactons on Evoltonary omptaton, vol. 0, no. 6, pp , [6] Y. G. Woldesenbet, G. G. Yen, and B. G. Tessema, onstrant handlng n mltobectve evoltonary optmzaton, IEEE Transactons on Evoltonary omptaton, vol. 3, no. 3, pp , [7] K. Deb, A. Pratap, S. Agarwal, and T. Meyarvan, A fast and eltst mltobectve genetc algorthm: NSGA-II, IEEE Transactons on Evoltonary omptaton, vol. 6, no. 2, pp , [8] A. Oyama, K. Shmoyama, and K. F New constrant-handlng method for mlt-obectve mlt-constrant evoltonary optmzaton and ts applcaton to space plane desgn, n Evoltonary and Determnstc Technqes for Desgn, Optmzaton and ontrol wth Applcatons to Indstral and Socetal Problems (EUROGEN 2005), Mnch, Germany, 2005, pp. -3. [9] D. hafekar, J. Xan, and K. Rasheed, onstraned mlt-obectve optmzaton sng steady state genetc algorthms, n Proc. Genetc and Evol. ompt. onf., 2003, pp [0] N. Yong, Blended rankng to cross nfeasble regons n constraned mltobectve problems, n Proceedngs of Internatonal onference on omptonal Intellgnece for Modelng, ontrol and Atomaton, and Internatonal onference on Intellgent Agents, Web Technologes and Internet ommerce, 2005, pp [] F. Jmenez, A. F. Gomez-Skarmeta, G. Sanchez, and K. Deb, An evoltonary algorthm for constraned mlt-obectve optmzaton, n Proc. E, 2002, pp [2] L. V. Santana-Qntero, A. G. Hernández-Díaz, J. Molna,. A. oello oello, and R. aballero, DEMORS: A hybrd mlt-obectve optmzaton algorthm sng dfferental evolton and rogh set theory for constraned problems, ompters & Operatons Research, vol. 37, no. 3, pp , 200. [3] H. K. Sngh, T. Ray, and W. Smth -PSA: onstraned Pareto smlated annealng for constraned mlt-obectve optmzaton, Informaton Scences, vol. 80, no. 3, pp , 200. [4] L. Jao, J. Lo, R. Shang, and F. L, A modfed obectve fncton method wth feasble-gdng strategy to solve constraned mltobectve optmzaton problems, Appled Soft omptng, vol. 4, pp , 204. [5] M. A. Jan and R. A. Khanm, A stdy of two penalty-parameterless constrant handlng technqes n the framework of MOEA/D, Appled Soft omptng, vol. 3, no., pp , 203. [6] T. P. Rnarsson and X. Yao, Stochastc rankng for constraned evoltonary optmzaton, IEEE Transactons on Evoltonary omptaton, vol. 4, no. 3, pp , [7] S. Qan, Y. Ye, B. Jang, and J. Wang, onstraned mltobectve optmzaton algorthm based on mmne system model, IEEE Transactons on ybernetcs, 205, n press. [8] B. Y. Q and P. N. Sganthan, onstraned mlt-obectve optmzaton algorthm wth an ensemble of constrant handlng methods, Engneerng Optmzaton, vol. 43, no. 4, pp , 20. [9] T. Takahama and S. Saka, onstraned optmzaton by the ε constraned dfferental evolton wth gradent-based mtaton and feasble eltes, n Proc. E, 2006, pp. -8. [20] K. Deb, A. Pratap, and T. Meyarvan, onstraned test problems for mlt-obectve evoltonary optmzaton, n Proceedngs of the Frst Internatonal onference on Evoltonary Mlt-rteron Optmzaton, 200, pp [2] Y. Wang, Z. a, Y. Zho, and W. Zeng. An adaptve tradeoff model for constraned evoltonary optmzaton, IEEE Transactons on Evoltonary omptaton, vol. 2, no., pp , [22] N. Noman and H. Iba, Acceleratng dfferental evolton sng an adaptve local search, IEEE Transactons on Evoltonary omptaton, vol. 2, no., pp , [23] E. Ztzler, L. Thele, M. Lamanns,. M. Fonseca, and V. G. da Fonseca, Performance assessment of mltobectve optmzers: An analyss and revew, IEEE Transactons on Evoltonary omptaton, vol. 7, no. 2, pp. 7-32, 2003.

Modeling Local Uncertainty accounting for Uncertainty in the Data

Modeling Local Uncertainty accounting for Uncertainty in the Data Modelng Local Uncertanty accontng for Uncertanty n the Data Olena Babak and Clayton V Detsch Consder the problem of estmaton at an nsampled locaton sng srrondng samples The standard approach to ths problem

More information

Scheduling with Integer Time Budgeting for Low-Power Optimization

Scheduling with Integer Time Budgeting for Low-Power Optimization Schedlng wth Integer Tme Bdgetng for Low-Power Optmzaton We Jang, Zhr Zhang, Modrag Potkonjak and Jason Cong Compter Scence Department Unversty of Calforna, Los Angeles Spported by NSF, SRC. Otlne Introdcton

More information

Numerical Solution of Deformation Equations. in Homotopy Analysis Method

Numerical Solution of Deformation Equations. in Homotopy Analysis Method Appled Mathematcal Scences, Vol. 6, 2012, no. 8, 357 367 Nmercal Solton of Deformaton Eqatons n Homotopy Analyss Method J. Izadan and M. MohammadzadeAttar Department of Mathematcs, Faclty of Scences, Mashhad

More information

Meta-heuristics for Multidimensional Knapsack Problems

Meta-heuristics for Multidimensional Knapsack Problems 2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,

More information

A New Approach For the Ranking of Fuzzy Sets With Different Heights

A New Approach For the Ranking of Fuzzy Sets With Different Heights New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

An Improved Isogeometric Analysis Using the Lagrange Multiplier Method

An Improved Isogeometric Analysis Using the Lagrange Multiplier Method An Improved Isogeometrc Analyss Usng the Lagrange Mltpler Method N. Valzadeh 1, S. Sh. Ghorash 2, S. Mohammad 3, S. Shojaee 1, H. Ghasemzadeh 2 1 Department of Cvl Engneerng, Unversty of Kerman, Kerman,

More information

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

PARETO BAYESIAN OPTIMIZATION ALGORITHM FOR THE MULTIOBJECTIVE 0/1 KNAPSACK PROBLEM

PARETO BAYESIAN OPTIMIZATION ALGORITHM FOR THE MULTIOBJECTIVE 0/1 KNAPSACK PROBLEM PARETO BAYESIAN OPTIMIZATION ALGORITHM FOR THE MULTIOBJECTIVE 0/ KNAPSACK PROBLEM Josef Schwarz Jří Očenáše Brno Unversty of Technology Faculty of Engneerng and Computer Scence Department of Computer Scence

More information

NGPM -- A NSGA-II Program in Matlab

NGPM -- A NSGA-II Program in Matlab Verson 1.4 LIN Song Aerospace Structural Dynamcs Research Laboratory College of Astronautcs, Northwestern Polytechncal Unversty, Chna Emal: lsssswc@163.com 2011-07-26 Contents Contents... 1. Introducton...

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

Multi-objective Optimization Using Adaptive Explicit Non-Dominated Region Sampling

Multi-objective Optimization Using Adaptive Explicit Non-Dominated Region Sampling 11 th World Congress on Structural and Multdscplnary Optmsaton 07 th -12 th, June 2015, Sydney Australa Mult-objectve Optmzaton Usng Adaptve Explct Non-Domnated Regon Samplng Anrban Basudhar Lvermore Software

More information

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique //00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy

More information

A General Algorithm for Computing Distance Transforms in Linear Time

A General Algorithm for Computing Distance Transforms in Linear Time Ths chapter has been pblshed as: A. Mejster, J. B. T. M. Roerdnk and W. H. Hesselnk, A general algorthm for comptng dstance transforms n lnear tme. In: Mathematcal Morphology and ts Applcatons to Image

More information

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming CS 4/560 Desgn and Analyss of Algorthms Kent State Unversty Dept. of Math & Computer Scence LECT-6 Dynamc Programmng 2 Dynamc Programmng Dynamc Programmng, lke the dvde-and-conquer method, solves problems

More information

Hybrid Method of Biomedical Image Segmentation

Hybrid Method of Biomedical Image Segmentation Hybrd Method of Bomedcal Image Segmentaton Mng Hng Hng Department of Electrcal Engneerng and Compter Scence, Case Western Reserve Unversty, Cleveland, OH, Emal: mxh8@case.ed Abstract In ths paper we present

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

OBJECT TRACKING BY ADAPTIVE MEAN SHIFT WITH KERNEL BASED CENTROID METHOD

OBJECT TRACKING BY ADAPTIVE MEAN SHIFT WITH KERNEL BASED CENTROID METHOD ISSN : 0973-739 Vol. 3, No., Janary-Jne 202, pp. 39-42 OBJECT TRACKING BY ADAPTIVE MEAN SHIFT WITH KERNEL BASED CENTROID METHOD Rahl Mshra, Mahesh K. Chohan 2, and Dhraj Ntnawwre 3,2,3 Department of Electroncs,

More information

Intra-Parametric Analysis of a Fuzzy MOLP

Intra-Parametric Analysis of a Fuzzy MOLP Intra-Parametrc Analyss of a Fuzzy MOLP a MIAO-LING WANG a Department of Industral Engneerng and Management a Mnghsn Insttute of Technology and Hsnchu Tawan, ROC b HSIAO-FAN WANG b Insttute of Industral

More information

Performance Modeling of Web-based Software Systems with Subspace Identification

Performance Modeling of Web-based Software Systems with Subspace Identification Acta Poltechnca Hngarca Vol. 13, No. 7, 2016 Performance Modelng of Web-based Software Sstems wth Sbspace Identfcaton Ágnes Bogárd-Mészöl, András Rövd, Shohe Yokoama Department of Atomaton and Appled Informatcs,

More information

Virtual Machine Migration based on Trust Measurement of Computer Node

Virtual Machine Migration based on Trust Measurement of Computer Node Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

5 The Primal-Dual Method

5 The Primal-Dual Method 5 The Prmal-Dual Method Orgnally desgned as a method for solvng lnear programs, where t reduces weghted optmzaton problems to smpler combnatoral ones, the prmal-dual method (PDM) has receved much attenton

More information

Multi-objective Virtual Machine Placement for Load Balancing

Multi-objective Virtual Machine Placement for Load Balancing Mult-obectve Vrtual Machne Placement for Load Balancng Feng FANG and Bn-Bn Qu,a School of Computer Scence & Technology, Huazhong Unversty Of Scence And Technology, Wuhan, Chna Abstract. The vrtual machne

More information

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints TPL-ware Dsplacement-drven Detaled Placement Refnement wth Colorng Constrants Tao Ln Iowa State Unversty tln@astate.edu Chrs Chu Iowa State Unversty cnchu@astate.edu BSTRCT To mnmze the effect of process

More information

Available online at Available online at Advanced in Control Engineering and Information Science

Available online at   Available online at   Advanced in Control Engineering and Information Science Avalable onlne at wwwscencedrectcom Avalable onlne at wwwscencedrectcom Proceda Proceda Engneerng Engneerng 00 (2011) 15000 000 (2011) 1642 1646 Proceda Engneerng wwwelsevercom/locate/proceda Advanced

More information

The Research of Support Vector Machine in Agricultural Data Classification

The Research of Support Vector Machine in Agricultural Data Classification The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou

More information

Clustering Algorithm Combining CPSO with K-Means Chunqin Gu 1, a, Qian Tao 2, b

Clustering Algorithm Combining CPSO with K-Means Chunqin Gu 1, a, Qian Tao 2, b Internatonal Conference on Advances n Mechancal Engneerng and Industral Informatcs (AMEII 05) Clusterng Algorthm Combnng CPSO wth K-Means Chunqn Gu, a, Qan Tao, b Department of Informaton Scence, Zhongka

More information

MODELING USER INTERESTS USING TOPIC MODEL

MODELING USER INTERESTS USING TOPIC MODEL Jornal of heoretcal and Appled Informaton echnology 0 th Febrary 203. Vol. 48 No. 2005-203 JAI & LLS. All rghts reserved. ISSN: 992-8645 www.att.org E-ISSN: 87-395 MODELING USER INERESS USING OPIC MODEL

More information

Problem Set 3 Solutions

Problem Set 3 Solutions Introducton to Algorthms October 4, 2002 Massachusetts Insttute of Technology 6046J/18410J Professors Erk Demane and Shaf Goldwasser Handout 14 Problem Set 3 Solutons (Exercses were not to be turned n,

More information

Study of Data Stream Clustering Based on Bio-inspired Model

Study of Data Stream Clustering Based on Bio-inspired Model , pp.412-418 http://dx.do.org/10.14257/astl.2014.53.86 Study of Data Stream lusterng Based on Bo-nspred Model Yngme L, Mn L, Jngbo Shao, Gaoyang Wang ollege of omputer Scence and Informaton Engneerng,

More information

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following.

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following. Complex Numbers The last topc n ths secton s not really related to most of what we ve done n ths chapter, although t s somewhat related to the radcals secton as we wll see. We also won t need the materal

More information

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS Academc Research Internatonal ISS-L: 3-9553, ISS: 3-9944 Vol., o. 3, May 0 EVALUATIO OF THE PERFORMACES OF ARTIFICIAL BEE COLOY AD IVASIVE WEED OPTIMIZATIO ALGORITHMS O THE MODIFIED BECHMARK FUCTIOS Dlay

More information

GIS . ( NSGA-II (GA NSGA-II) GIS

GIS . ( NSGA-II (GA NSGA-II)   GIS Vol9, No 3, Autumn 07 Iranan Remote Sensng & 5-3 3 * 3 /6/5 /5/ 35 9 * Emal zdarvar@gmalcom - 60 50 05 3 387 Datta and Deb, Deb et al, 005 388 Herzng, 008 Datta et al, 007 006 Deb and Sundar, 006 Vllalta-Calderon

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

Obstacle Avoidance by Using Modified Hopfield Neural Network

Obstacle Avoidance by Using Modified Hopfield Neural Network bstacle Avodance by Usng Modfed Hopfeld Neral Network Panrasee Rtthpravat Center of peraton for Feld Robotcs Development (FIB), Kng Mongkt s Unversty of Technology Thonbr. 91 Sksawas road Tongkr Bangkok

More information

Restaurants Review Star Prediction for Yelp Dataset

Restaurants Review Star Prediction for Yelp Dataset Restarants Revew Star Predcton for Yelp Dataset Mengq Y UC San Dego A53077101 mey004@eng.csd.ed Meng Xe UC San Dego A53070417 m6xe@eng.csd.ed Wenja Oyang UC San Dego A11069530 weoyang@eng.csd.ed ABSTRACT

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

Type-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data

Type-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data Malaysan Journal of Mathematcal Scences 11(S) Aprl : 35 46 (2017) Specal Issue: The 2nd Internatonal Conference and Workshop on Mathematcal Analyss (ICWOMA 2016) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES

More information

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

A combined test for randomness of spatial distribution of composite microstructures

A combined test for randomness of spatial distribution of composite microstructures ISSN 57-7076 Revsta Matéra, v., n. 4, pp. 597 60, 007 http://www.matera.coppe.frj.br/sarra/artgos/artgo0886 A combned test for randomness of spatal dstrbton of composte mcrostrctres ABSTRACT João Domngos

More information

Support Vector Machines

Support Vector Machines Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned

More information

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal

More information

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1) Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A

More information

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

Learning-Based Top-N Selection Query Evaluation over Relational Databases

Learning-Based Top-N Selection Query Evaluation over Relational Databases Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **

More information

J1.8 APPLICATION OF CFD SIMULATIONS FOR SHORT-RANGE ATMOSPHERIC DISPERSION OVER OPEN FIELDS AND WITHIN ARRAYS OF BUILDINGS

J1.8 APPLICATION OF CFD SIMULATIONS FOR SHORT-RANGE ATMOSPHERIC DISPERSION OVER OPEN FIELDS AND WITHIN ARRAYS OF BUILDINGS AMS th Jont Conference on the Applcatons of Ar Pollton Meteorology wth the A&WMA, Atlanta, GA, Jan - Feb, 6. J.8 APPLICATION OF CFD SIMULATIONS FOR SHORT-RANGE ATMOSPHERIC DISPERSION OVER OPEN FIELDS AND

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and

More information

An Optimal Algorithm to Find a Minimum 2-neighbourhood Covering Set on Cactus Graphs

An Optimal Algorithm to Find a Minimum 2-neighbourhood Covering Set on Cactus Graphs Annals of Pre Appled Mathematcs Vol 2 No 1 212 45-59 ISSN: 2279-87X (P) 2279-888(onlne) Pblshed on 18 December 212 wwwresearchmathscorg Annals of An Optmal Algorthm to Fnd a Mnmm 2-neghborhood overng Set

More information

Study on Multi-objective Flexible Job-shop Scheduling Problem considering Energy Consumption

Study on Multi-objective Flexible Job-shop Scheduling Problem considering Energy Consumption Journal of Industral Engneerng and Management JIEM, 2014 7(3): 589-604 nlne ISSN: 2014-0953 Prnt ISSN: 2014-8423 http://dx.do.org/10.3926/jem.1075 Study on Mult-objectve Flexble Job-shop Schedulng Problem

More information

Performance Evaluation of Information Retrieval Systems

Performance Evaluation of Information Retrieval Systems Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence

More information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION 24 CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION The present chapter proposes an IPSO approach for multprocessor task schedulng problem wth two classfcatons, namely, statc ndependent tasks and

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

Reducing Frame Rate for Object Tracking

Reducing Frame Rate for Object Tracking Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg

More information

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr) Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute

More information

Wishing you all a Total Quality New Year!

Wishing you all a Total Quality New Year! Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma

More information

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation Internatonal Conference on Logstcs Engneerng, Management and Computer Scence (LEMCS 5) Maxmum Varance Combned wth Adaptve Genetc Algorthm for Infrared Image Segmentaton Huxuan Fu College of Automaton Harbn

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

Boundary layer and mesh refinement effects on aerodynamic performances of horizontal axis wind turbine (HAWT)

Boundary layer and mesh refinement effects on aerodynamic performances of horizontal axis wind turbine (HAWT) Bondary layer and mesh refnement effects on aerodynamc performances of horzontal axs wnd trbne (HAWT) YOUNES EL KHCHINE, MOHAMMED SRITI Engneerng Scences Laboratory, Polydscplnary Faclty Sd Mohamed Ben

More information

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster

More information

Load-Balanced Anycast Routing

Load-Balanced Anycast Routing Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance

More information

Classification / Regression Support Vector Machines

Classification / Regression Support Vector Machines Classfcaton / Regresson Support Vector Machnes Jeff Howbert Introducton to Machne Learnng Wnter 04 Topcs SVM classfers for lnearly separable classes SVM classfers for non-lnearly separable classes SVM

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks Seventh Internatonal Conference on Intellgent Systems Desgn and Applcatons GA-Based Learnng Algorthms to Identfy Fuzzy Rules for Fuzzy Neural Networks K Almejall, K Dahal, Member IEEE, and A Hossan, Member

More information

A fast algorithm for color image segmentation

A fast algorithm for color image segmentation Unersty of Wollongong Research Onlne Faculty of Informatcs - Papers (Arche) Faculty of Engneerng and Informaton Scences 006 A fast algorthm for color mage segmentaton L. Dong Unersty of Wollongong, lju@uow.edu.au

More information

12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification

12/2/2009. Announcements. Parametric / Non-parametric. Case-Based Reasoning. Nearest-Neighbor on Images. Nearest-Neighbor Classification Introducton to Artfcal Intellgence V22.0472-001 Fall 2009 Lecture 24: Nearest-Neghbors & Support Vector Machnes Rob Fergus Dept of Computer Scence, Courant Insttute, NYU Sldes from Danel Yeung, John DeNero

More information

Fast Computation of Shortest Path for Visiting Segments in the Plane

Fast Computation of Shortest Path for Visiting Segments in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 4 The Open Cybernetcs & Systemcs Journal, 04, 8, 4-9 Open Access Fast Computaton of Shortest Path for Vstng Segments n the Plane Ljuan Wang,, Bo Jang

More information

Journal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data

Journal of Chemical and Pharmaceutical Research, 2014, 6(6): Research Article. A selective ensemble classification method on microarray data Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(6):2860-2866 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A selectve ensemble classfcaton method on mcroarray

More information

Hierarchical clustering for gene expression data analysis

Hierarchical clustering for gene expression data analysis Herarchcal clusterng for gene expresson data analyss Gorgo Valentn e-mal: valentn@ds.unm.t Clusterng of Mcroarray Data. Clusterng of gene expresson profles (rows) => dscovery of co-regulated and functonally

More information

Module Management Tool in Software Development Organizations

Module Management Tool in Software Development Organizations Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,

More information

Multi-objective Optimization Using Self-adaptive Differential Evolution Algorithm

Multi-objective Optimization Using Self-adaptive Differential Evolution Algorithm Mult-objectve Optmzaton Usng Self-adaptve Dfferental Evoluton Algorthm V. L. Huang, S. Z. Zhao, R. Mallpedd and P. N. Suganthan Abstract - In ths paper, we propose a Multobjectve Self-adaptve Dfferental

More information

Cost-efficient deployment of distributed software services

Cost-efficient deployment of distributed software services 1/30 Cost-effcent deployment of dstrbuted software servces csorba@tem.ntnu.no 2/30 Short ntroducton & contents Cost-effcent deployment of dstrbuted software servces Cost functons Bo-nspred decentralzed

More information

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures

A Novel Adaptive Descriptor Algorithm for Ternary Pattern Textures A Novel Adaptve Descrptor Algorthm for Ternary Pattern Textures Fahuan Hu 1,2, Guopng Lu 1 *, Zengwen Dong 1 1.School of Mechancal & Electrcal Engneerng, Nanchang Unversty, Nanchang, 330031, Chna; 2. School

More information

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation Intellgent Informaton Management, 013, 5, 191-195 Publshed Onlne November 013 (http://www.scrp.org/journal/m) http://dx.do.org/10.36/m.013.5601 Qualty Improvement Algorthm for Tetrahedral Mesh Based on

More information

Clustering Algorithm of Similarity Segmentation based on Point Sorting

Clustering Algorithm of Similarity Segmentation based on Point Sorting Internatonal onference on Logstcs Engneerng, Management and omputer Scence (LEMS 2015) lusterng Algorthm of Smlarty Segmentaton based on Pont Sortng Hanbng L, Yan Wang*, Lan Huang, Mngda L, Yng Sun, Hanyuan

More information

Collaboratively Regularized Nearest Points for Set Based Recognition

Collaboratively Regularized Nearest Points for Set Based Recognition Academc Center for Computng and Meda Studes, Kyoto Unversty Collaboratvely Regularzed Nearest Ponts for Set Based Recognton Yang Wu, Mchhko Mnoh, Masayuk Mukunok Kyoto Unversty 9/1/013 BMVC 013 @ Brstol,

More information

Video Proxy System for a Large-scale VOD System (DINA)

Video Proxy System for a Large-scale VOD System (DINA) Vdeo Proxy System for a Large-scale VOD System (DINA) KWUN-CHUNG CHAN #, KWOK-WAI CHEUNG *# #Department of Informaton Engneerng *Centre of Innovaton and Technology The Chnese Unversty of Hong Kong SHATIN,

More information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

Evolutionary Multi-Objective Optimization for Mesh Simplification of 3D Open Models

Evolutionary Multi-Objective Optimization for Mesh Simplification of 3D Open Models Evolutonary Mult-Objectve Optmzaton for Mesh Smplfcaton of 3D Open Models B. Rosaro Campomanes-Álvarez a,*, Oscar Cordón abc and Sergo Damas a a European Centre for Soft Computng, Gonzalo Gutérrez Qurós

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

More information

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between

More information

Programming in Fortran 90 : 2017/2018

Programming in Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Exercse 1 : Evaluaton of functon dependng on nput Wrte a program who evaluate the functon f (x,y) for any two user specfed values

More information

An Indian Journal FULL PAPER ABSTRACT KEYWORDS. Trade Science Inc.

An Indian Journal FULL PAPER ABSTRACT KEYWORDS. Trade Science Inc. [Type text] [Type text] [Type text] ISSN : 97-735 Volume Issue 9 BoTechnology An Indan Journal FULL PAPER BTAIJ, (9), [333-3] Matlab mult-dmensonal model-based - 3 Chnese football assocaton super league

More information

A Similarity-Based Prognostics Approach for Remaining Useful Life Estimation of Engineered Systems

A Similarity-Based Prognostics Approach for Remaining Useful Life Estimation of Engineered Systems 2008 INTERNATIONAL CONFERENCE ON PROGNOSTICS AND HEALTH MANAGEMENT A Smlarty-Based Prognostcs Approach for Remanng Useful Lfe Estmaton of Engneered Systems Tany Wang, Janbo Yu, Davd Segel, and Jay Lee

More information

A Clustering Algorithm for Chinese Adjectives and Nouns 1

A Clustering Algorithm for Chinese Adjectives and Nouns 1 Clusterng lgorthm for Chnese dectves and ouns Yang Wen, Chunfa Yuan, Changnng Huang 2 State Key aboratory of Intellgent Technology and System Deptartment of Computer Scence & Technology, Tsnghua Unversty,

More information

An Adaptive Virtual Machine Location Selection Mechanism in Distributed Cloud

An Adaptive Virtual Machine Location Selection Mechanism in Distributed Cloud KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS VOL. 9, NO. 12, Dec. 2015 4776 Copyrght c2015 KSII An Adaptve Vrtual Machne Locaton Selecton Mechansm n Dstrbuted Cloud Shukun Lu 1, Wea Ja 2 1 School

More information

An Improved Image Segmentation Algorithm Based on the Otsu Method

An Improved Image Segmentation Algorithm Based on the Otsu Method 3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,

More information

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1

A New Feature of Uniformity of Image Texture Directions Coinciding with the Human Eyes Perception 1 A New Feature of Unformty of Image Texture Drectons Concdng wth the Human Eyes Percepton Xng-Jan He, De-Shuang Huang, Yue Zhang, Tat-Mng Lo 2, and Mchael R. Lyu 3 Intellgent Computng Lab, Insttute of Intellgent

More information

Design and Analysis of Algorithms

Design and Analysis of Algorithms Desgn and Analyss of Algorthms Heaps and Heapsort Reference: CLRS Chapter 6 Topcs: Heaps Heapsort Prorty queue Huo Hongwe Recap and overvew The story so far... Inserton sort runnng tme of Θ(n 2 ); sorts

More information

Study on Properties of Traffic Flow on Bus Transport Networks

Study on Properties of Traffic Flow on Bus Transport Networks Study on Propertes of Traffc Flow on Bus Transport Networks XU-HUA YANG, JIU-QIANG ZHAO, GUANG CHEN, AND YOU-YU DONG College of Computer Scence and Technology Zhejang Unversty of Technology Hangzhou, 310023,

More information

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009. Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton

More information