An Optimization Algorithm Based on Binary Difference and Gravitational Evolution

Size: px
Start display at page:

Download "An Optimization Algorithm Based on Binary Difference and Gravitational Evolution"

Transcription

1 Internatonal Journal of Computatonal Intellgence Systems Vol. 5 No. 3 (June 0) An Optmzaton Algorthm Based on Bnary fference and Gravtatonal Evoluton Junl L Yang Lou and Yuhu Sh 3. College of Computer Scence Schuan Normal Unversty Chengdu Chna. Informaton Scence and Engneerng College Nngbo Unversty Nngbo 35 Chna 3. epartment of Electrcal and Electronc Engneerng X an Jaotong-Lverpool Unversty Suzhou 53 Chna E-mal: l.unl@vp.63.com louyang@mal.nbu.edu.cn Yuhu.Sh@xtlu.edu.cn Abstract Unversal gravtaton s a natural phenomenon. Inspred by Newton's unversal gravtaton model and based on bnary dfferences strategy we propose an algorthm for global optmzaton problems whch s called the bnary dfference gravtatonal evoluton (BGE) algorthm. BGE s a populaton-based algorthm and the populaton s composed of partcles. Each partcle s treated as a vrtual obect wth two attrbutes of poston and qualty. Some of the best obects n the populaton compose the reference-group and the rest obects compose the floatng-group. The BGE algorthm could fnd the global optmum solutons through two crtcal operatons: the self-update of reference-group and the nteractve-update process between the reference-group and floatng-group utlzng the gravtatonal evoluton method. The parameters of BGE are set by a tral-and-error process and the BGE s proved that t can converge to the global optmal soluton wth probablty. Benchmar functons are used to evaluate the performance of BGE and to compare t wth classc fferental Evoluton. The smulaton results llustrate the encouragng performance of the BGE algorthm wth regards to computng speed and accuracy. Keywords: Optmzaton; Bnary fference; fferental Evoluton; Gravtaton.. Introducton Evolutonary algorthms are a seres of problem-solvng methods that based on smulaton of the natural evoluton system and ts development can be traced bac to 950s. Compared wth the classc optmzaton methods an evolutonary algorthm has many advantages. For example t s unconstraned by the search space lmtatons; t s unconstraned by the functon types; and functon gradent nformaton s not essental etc. Evolutonary algorthms have been wdely used for optmzaton problem. Bologcal evoluton theory nspred the emergence and development of boncs whch motvated many types of evolutonary algorthms such as Genetc Algorthm [-] Smulated Annealng [3] Immune Algorthm [-6] Partcle Swarm Optmzaton [7-9] Ant Colony Optmzaton [0-] and Bacteral Foragng Optmzaton [] etc. All these contrbutons have ganed great achevements on the feld of optmzaton. Chrs and Tsang frstly proposed the concept and frame of GELS (Gravtatonal Emulaton Local Search) [3] n 995 whch was then further developed by Webster []. Balachandar and Kannan [5] proposed RGES (Randomzed Gravtatonal Emulaton Search) algorthm n 007 whch overcame some wea ponts of GELS such as a relatvely slow convergence rate and low qualty of solutons. The three gravtatonal emulaton search methods above were ntally appled to solve combnatoral optmzaton problems such as Travelng Salesman Problem (TSP). Hsao and Chuang et. al. proposed the SGO (Space 83

2 J. L et al. Gravtatonal Optmzaton) [6] whch was based on Ensten s theory and Newton s law of gravtaton. The SGO smulated the process that a number of planets shft n the space to search for the planet wth the most massve. The geometrc transformaton of the space generates a force to mae the planets shftng faster or slower. In 007 Chuang and Jang proposed an algorthm named IRO (Integrated Radaton Optmzaton) [7] based on the phenomenon that the movements of one planet n the gravtatonal feld were compostvely effected by the sum of all the other planets gravtatonal radaton forces. Rashed proposed the GSA (Gravtatonal Search Algorthm) n 007 and t was contnuously mproved thereafter [8-0]. In GSA a partcle s total gravtaton was a sum of all other partcles gravtatons wth random weghts and the total gravtaton generated the acceleraton to move. GPSO (Partcle Swarm Optmzaton based on Gravtaton) [] was proposed by Kang and Wang et. al n 007 whch ntroduced an acceleraton nto Partcle Swarm Optmzaton. All the four algorthms above were methods based on the poston and dsplacement and nspred by gravtaton each of whch has dfferent update formulas respectvely. In ths paper the proposed algorthm Bnary- fference Gravtatonal Evoluton (BGE) shares the same pont wth the algorthms above that s all these algorthms utlze the concept of gravtaton. However BGE does not smulate the physcal movement processes as above but va a clusterng process based on eltst strategy. In BGE obects n the populaton are clustered nto two dfferent groups: the referenceobect group and the floatng-obect group. Obects of dfferent groups are updated ndependently or cooperatvely and the bnary dfference [5] strategy s adopted to update obects n the updatng processes ndependently or cooperatvely.e. the self-update of reference-group and the nteractve-update group and nteractng process between the two groups. Ths paper s organzed as follows. Secton presents both the detals and ensemble of BGE. Expermental study s gven n Secton 3 where benchmar functons are used to evaluate the performance. Secton gves an analyss on the parameters and the convergence of algorthm s analyzed n Secton 5 followed by the concluson n Secton 6.. Bnary fference Gravtatonal Evoluton Algorthm.. Bnary fference Strategy The fferental Evoluton (E) [-3] algorthm emerged as a very compettve form of evolutonary algorthm more than a decade ago and was frst proposed by R. Storn and K. Prce n 995 whch was a smple and effcent heurstc algorthm for global optmzaton over contnuous spaces and ts feature s a mutaton wth a dfferental strategy. The selecton of dfferental strateges would mae sgnfcant nfluence on the performance of the algorthm. In consderaton of dversty there are at least three ndvduals nvolved n mutaton n E [3-]. To smplfy the mutaton operaton the bnary dfference strategy was ntroduced whch s smlar to the tradtonal but wth only two ndvduals nvolved. As dscussed n [5] bnary dfference usng a sorted populaton could mprove the performance of E especally n the aspect of convergence speed for low-dmensonal problems. Because the obects of the reference-obect group n BGE are sorted by ther qualtes bnary dfference s approprate to be ntroduced nto BGE. In bnary dfference two obects are selected as canddates of whch the one wth better ftness value s selected as the central obect and the obects wth relatve worse ftness values are nvolved n mutaton. The process of producng new obects s as follows. A central obect s temporarly fxed and the worse-ftness obects are updated one after another accordng to the central obect. The new obects are more lely to be close to the central obect. The bnary dfference strategy not only s smpler but also taes advantages of the gravtaton clusterng. The populaton s dvded nto two groups by gravtaton clusterng and the bnary dfference s used for the nteractve update of the two groups. The dmensonal update of an obect wth bnary dfference strategy s as follows: X ( new ) X ( ) csgn c U(0) X ( ) X ( ) where X () and X( ) are two obects n a populaton and X ( new) represents the newly produced ndvdual. The subscrpt means dmenson csgn s a random symbol c s a constant and U (0) s a random value unformly dstrbuted n [0]. Bnary dfference uses a 8

3 Bnary fference Gravtatonal Evoluton random and varable factor csgn c U(0) to eep dversty of the populaton... Gravtatonal Groupng Mode Gravtaton s an attractve force between two obects whch exsts between any two obects wth masses. The value of gravtaton s n proporton to the mass of ether obect whle n nverse proporton to the dstance between them. Calculaton of gravtaton can be descrbed as follows: m m F G () r where G 6.67 s the gravtatonal constant m and m are qualtes of two obects and r represents the dstance between two obects. Wthout loss of generalty we consder only the maxmzaton problem as optmzaton problem consdered n ths paper: Z max f ( X ) XS The search space s S { X l x u... n} where X ( x x x ) T n. l u( n) are the lower and upper bound. f () s the obectve functon. We treat a partcle n the search space S as an obect usng gravtaton for optmzaton. For any two obects A and A n the space S ther physcal postons are represented by X( A ) and X( A ) and ther qualtes are represented by ma ( ) and ma ( ). Then the gravtaton of A and A s smply descrbed as: F m( A ) m( A ) h( f ( A )) h( f ( A )) r K r K 0 0 where r means dstance measurement and K0 s a small-valued constant to ensure denomnator unequal to zero n Formula (). h() s a one-dmensonal scale () transformaton functon satsfed x ( ) hx ( ) 0 whch strctly ncrease monotoncally. For convenence h() s set as the absolute value here n accordance wth the concept that qualty s a nonnegatve value. m( ) h( f ( )) represents the qualty of an obect whch s scale transformed from obectve functon. Formula () s essentally equal to Formula () though the form s changed. In ths paper we calculate the gravtaton measurements but not the true gravtaton values. For convenence the ftness of an obect we mentoned n ths paper ndcates the qualty whch s a scale transformaton of the obectve functon..3. Bnary fference Gravtatonal Evoluton Summary The populaton denoted by RN s assemble of all obects whch are clustered nto two groups.e. the reference-obect group R and the floatng-obect group N. Followngs are the symbols used n the BGE descrptons. X( R ) represents the poston of the th reference-obect R X( R ) represents ts th dmenson and mr ( ) represents the qualty of the reference-obect R. Smlarly X( N ) represents the poston X( N ) represents ts th dmenson and mn ( ) represents the qualty of the floatng-obect N. U( a b) s a random value unformly dstrbuted n nterval [ ab. ] Bnary fference Gravtatonal Evoluton algorthm can be summarzed as follows: Step The reference-obect group R and the floatng-obect group N are randomly generated: R { R n} N { N n} where n and n are two nteger numbers and represent the group szes of two groups respectvely. Step If the halt condtons are satsfed then halt. Otherwse resort the populaton RN whch should satsfy the followng rules: m( R ) m( R ) n m( R) m( N ) n n Step3 Bnary dfference strategy s appled to any two reference-obects R ( n ) and R ( n ) and a new obect C / ( n) s generated: 85

4 J. L et al. U (0) 0.5 csgn n U (0) 0.5 U ( l u ) f X ( R ) X ( R ) XC ( / ) X ( R ) csgn c U (0) X ( R ) X ( R ) otherwse where XC ( / ) XC ( / ) and mc ( / ) represent the poston the th dmenson and the qualty of the newly generated obect C / respectvely. X ( R ) X ( R ) represents the dstance of two obects. C / replaces Rn f m( C ) m( R ) and then / resort the reference-obect group R whch should satsfy the followng rules: n m( R ) m( R ) n ; Step The gravtaton measurement F/ and dstance measurement r / (Eucldean dstance measurement s used here) between any R ( n ) and N ( n ) s calculated as follows: r X ( R ) X ( N ) m( R) m( N ) F/ r/ K0 / Then a bdrectonal selecton runs: () For every floatng-obect N ( n ) to select a reference-obect R ndex _( ) whch has the maxmal gravtaton wth N where max F / max( F / n ) max ndex _( ) mn({ F / F / n}) () For every reference-obect R ( n ) to select a floatng-obect N ndex _ ( ) whch has the mnmum gravtaton wth R ( n ) and then t s thorough elmnated whch means t would be replaced and do not partcpate n generatng the next generatons where set( ) { ndex _( ) n} mn F / mn({ F / set( )}) mn ndex _ ( ) mn({ F / F / set( )}) The elmnated obect N ndex _ ( ) does not exst f set(). Step5 There are two stuatons for update of floatng-obects based on step. () For the floatng-obects whch have not been elmnated { N ndex _( ) n n} bnary dfference s appled to update them: U (0) 0.5 csgn n U (0) 0.5 U ( l u ) f X ( Rndex _( ) ) X ( N ) X( Rndex _( ) ) X( N ) csgn c U (0) X ( Rndex _( ) ) X ( N ) otherwse () For the elmnated floatng-obects { N ndex _ ( ) n} we randomly select a reference-obect to replace t. Then go to Step... Algorthm Illustratons... Sortng Operaton The populaton s dvded nto the reference-obect group and the floatng-obect group n BGE where the reference-obects have the better ftness values thus they are obects wth bgger qualtes. The referenceobects are arranged accordng to ther qualtes n the group. Reference-obects Orderng Rules 3 m() m() m(3) m() Fg. Orderng of Reference-obect Group The sortng operaton s appled to the referenceobect group only and the rules of sortng s llustrated n Fgure where m() represent qualty of an obect and obects are sorted n a proper sequence by ther values of qualtes. The other part of populaton.e. the floatng-obect group does not need to be sorted and the only requrement we need to ensure s that any floatng-obect s ftness value should never be better than the worst reference-obect s. 86

5 Bnary fference Gravtatonal Evoluton... Method of Update There are two types of update methods n BGE.e. the self-update of reference-group and the nteractveupdate process between the reference-group and floatng-group. Reference-obects Update of Referenceobects Themselves C / C /3 C /5 C /6 Fg. An Example of Reference-obects Self-update As t s showed n Fgure the self-update process of the reference-obects occurred after obects are sorted. Reference-obects and and 3 and 3 are used to produce new obects one par after another by bnary dfference. Thus any two referenceobects are selected n the process above. The rule of new obects selecton s that we select the superor wth bgger qualty and elmnate the nferor. The purpose of a reference-obect s to eep a hgh level on qualty as well as ftness whch to a certan degree gnores the dversty of the reference-group but speed up the convergence rate only. regeneraton s called forced-update whch means beng replaced and not partcpate n generatng the next generatons. The forced-update step of nteractveupdate contrbutes a lot n the goal of eepng populaton dversty...3. Physcal Meanngs As showed n both Formula () and () f two obects have a large value of gravtaton that may be caused by ether a small dstance n the denomnator or bg qualty of each obect n the numerator. The second case s what we expected whle the former may lead to a trap of local optmum. To solve ths problem we ntroduce a threshold parameter the value of whch wll be dscussed n another paragraph.... Advantages of Gravtaton Clusterng: Reference-obects 3 Floatng-obect Fg. 3 An Example of Gravtaton Clusterng After the self-update process of reference-group an nteractve-update of reference-obects and floatngobects are performed. Fgure 3 shows an example of gravtaton clusterng the essental of whch s the nteractve-update. Compared wth reference-obects and 3 floatng-obect has a bgger gravtaton measurement wth reference-obect so floatngobect selects reference-obect the lnes of reference-obect and floatng-obect shows ther relatonshp. The smlar progress occurs on other obects. For the reference-obect there has four floatng-obects ( 3 and 8) selected among whch floatng-obect 3 (wth a dot) has the mnmum gravtaton measurement wth reference-obect then floatng-obect 3 would be elmnated from populaton. Smlarly other reference-obects elmnate floatngobects as above to fnsh the nteractve-update process between the reference-group and floatng-group. The postons of elmnated floatng-obects are regenerated n a random way as ntalzaton. The Fg. Searchng Method Fgure shows an example of one dmenson as shown reference -obects R R and floatng-obects N N accordng to Formula () F m( R ) m( N) / ( r K0) F m( R ) m( N) / ( r K0). Both F and F are large (when F s small the relatonshp of R and N s gnored) where F s caused by bg qualtes of R and t s good for explorng on new regon whch ncreases dversty. On the contrary F may be caused by the small dstance r. Ths s good for explotng n the current regon whch ncreases accuracy but may cause a rs of beng trapped n local optmums. 3. Experments 3.. Comparson Methods We choose fferental Evoluton to compare BGE wth because both use dfference vector to generate new ndvduals. The parameters of the two algorthms are set as follows: 87

6 J. L et al. E: The parameters are set as that proposed by Yang et al.[6] that s the populaton sze NP 0 where s the dmenson of varable n soluton space CR 0.9 F 0.5 and a mutaton strategy of E/rand/. The haltng condton s:. The number of ftness functon evaluatons overstepped Supposed Mnf s the global optmum value f best s the searched optmum value and 6 0 s the accuracy value f Mnf fbest the runnng stops. BGE: The reference-obect group sze n 0 the floatng-obect group sze n 0 c K 0 0 as that n E Benchmar Functons and 9.And 0. The haltng condton s set the same There are 9 benchmar functons presented n table whch are used to evaluate the performance of BGE and to compare t wth the classc fferental Evoluton. Table. Benchmar Functons Test Functon oman Range Optmal Pont Sphere Model [-0000] f x Mn f f(00 0) 0 Schwefel's Problem. f x x [-00] Mn f f (00 0) 0 Schwefel's Problem. f3 max{ x } [-0000] Mn f 3 f (00 0) 0 Step Functon f ( x 0.5 ) Quartc Functon.e. Nose f x random[0) 5 Generalzed Schwefel's Problem.6 f6 ( xsn( x )) Generalzed Rastrgn Functon 7 f x 0cos( x ) 0 Generalzed Penalzed Functon I f y y 8 {0sn ( ) ( ) [ 0sn ( y )] ( y ) } where ux ( 000 ) [-0000] Mn f f (00 0) 0 [-.8.8] Mn f 5 f (00 0) Mn f6 [ ] f ( ) [-5.5.] Mn f 7 f (00 0) 0 [-5050] Mn f 8 f ( ) m ( x a) x a u( x a m) 0 a x a m ( x a) x a y ( x ) Generalzed Penalzed Functon II 9 0.{sn (3 ) ( ) f x x [ sn (3 x )] ( x ) [ sn ( x)]} where ux ( 500) m ( x a) x a u( x a m) 0 a x a m ( x a) x a 3.3. Smulaton Results [-5050] Mn f 9 f ( ) 0 We run each algorthm 50 tmes ndependently on each benchmar functon above. The dmenson of problems s set to be 0. Successful run means the dfference between the obtaned soluton and the true value to be 6 less than 0 before the number of functon evaluatons reaches the maxmum value whch s set to be Test Functon Table. Comparson of results on 50-tme ndependent random testng wth benchmar functons BGE E Best/Worst Std FEs Best/Worst Std FEs f 3.77e-7/9.86e-6 e e-6/9.98e-6 e-6 0 f.79e-6/9.97e-6 e e-6/9.9e-6 e f3 0/ / f 0/ / f5.0e-6/3.9e- e e-5/.0e- e f / e / e f7.9e-9/9.83e-6 e e-6/0.99 e f8 5.8e-8/9.96e-6 e e-6/9.9e-6 e f9.e-9/9.93e-6 e e-6/9.98e-6 e-6 09 As shown n Table where Best /Worst mean the best and worst solutons n the testng based on the average of 50 tmes of ndependent runs; Std and FEs s short for standard devaton and functon evaluatons. Accordng to the data n Table fferental Evoluton cannot always fnd the optmal value of f 7 for the worst soluton n the 50 runs s dssatsfyng the precson. We defne when the obtaned soluton whch has the precson less then 0 as a successful run. There are 6 successful runs n 50 total runs whle 9 88

7 Bnary fference Gravtatonal Evoluton optmzng f 7 utlzng E whle for other functons n ths test the successful runs are all 50.e. E can solve the rest problems and the proposed BGE can optmze all the functons wth relatve smaller Std and FEs values whch means a hgher precson and fast convergence speed Table shows BGE has a better performance than E due to a smaller number of functon evaluatons of whch a maxmal dsparty was that E got a more than 0-tme functon evaluatons than BGE that means the latter s 0 tmes faster than the former. Fgure 5 shows the comparson between E and BGE whch llustrate the ftness values decreased as teraton ncreasng for optmzng the functons. (c) Fg. 5 Comparson of BGE and E n Testng of (a)f (b)f (c)f8. As can be observed clearly from the fgures BGE reached the global optmum requrng less number of teraton than E. urng the optmzng process the blue sold lne declnes faster than the red dotted lne whch llustrates that the BGE converges faster than E does.. Expermental Study of Parameters (a).. The group szes n and n The values of group sze n and n may nfluence the performance of BGE a lot. Here we dscuss these two parameters usng Schaffer Problem (SF) as an example whch was a classcal deceptve problem. sn x 0.5 f x SF 0.00( x ) Mn f f (00 0) 0 SF SF (b) We optmze f SF wth -dmenson and 5-dmenson utlzng BGE to test dfferent settngs of group sze n and n. The halt condton s set as ether f f SF 6 Mn 0 or the number of ftness functon evaluatons overstepped the Max-FEs whch was set to be for -dmenson problem and for 5-dmenson problem respectvely. Other parameters of 9 0 BGE are set as follows 0 c K 0 0. n and n are expermentally set to be

8 J. L et al. respectvely. For each par of n and n 50 runs are conducted ndependently and a total of tmes of runs are conducted for the two problems. Fgure 6 shows the nfluences on optmzaton f SF of -dmenson and Fgure. 7 shows that of 5-dmenson. Accordng to the halt condton we adopt success as 6 operaton stopped when f Mn f SF 0 other than the number of ftness functon exceeds Max-FEs. It can be seen from Fgure 6 that for a -dmenson problem when n n t has a hgher success rate whle a lower rate when n n. From Fgure 6(b) when n n t has a lower number of average functon evaluaton whle an opposte result when n n. As an observaton from Fgure 6 a selecton of n ns good for the -dmenson Schaffer Problem. The same observaton can be obtaned from Fgure 7 for the 5- dmenson problem. Intutvely accordng to the algorthm procedure when n n the reference-obects update taes hgh percentage of the whole update progress whch leads a fast convergence. However t wll lose dversty especally when dealng wth deceptve problem. When n n ths s another status that the dversty s contented enough but wth a lttle slow convergence. BGE s an algorthm that can balance between a hgh convergence rate and a hgh dversty. The floatng-obect group sze could be arbtrarly bg but a too large group sze s tme consumng. The concluson s n should be approprate small n could be slghtly larger than n and ther relatonshp should meet n n. The specfc values are defned dependng on crcumstances. (b) Fg. 6 The Effect of Alternatve Group Szes on Optmzaton Schaffer Problem (SF) of - dmenson (a) Result of Success Rate (b) Result of Average Functon Evaluatons (a) (b) Fg. 7 The Effect of Alternatve Group Szes on Optmzaton Schaffer Problem (SF) of 5- dmenson (a) Result of Success Rate (b) Result of Average Functon Evaluatons.. The threshold parameter (a) The threshold parameter s dscussed n ths part whch nfluences the algorthm performance a lot as well. BGE s tested on Shfted Sphere Functon ( f SSF ). 90

9 Bnary fference Gravtatonal Evoluton SSF f z z x o 00 x 00 u l 3 l ( u l ) o u l 3 l ( u l ) 3 Mn f f ( o o o ) 0 SSF SSF Here we separately test the problem wth a dmenson of and 00 respectvely. The parameters are 0 set as follows: n 0 n 0 c K 0 0. The halt condtons are set as ether f f SSF the number of ftness functon evaluatons overstepped ncreases gradually by Mn 0 or 0 0 and for each settng of 50 runs are conducted ndependently. Table. 3 The result of benchmar testng on Shfted Sphere Functon wth dfferent values of Test Shfted Sphere Functon functons Std FEs Std FEs Std FEs Std FEs 0 - e e e e e e e e e e e e e e e e e e e e e-8 69 e e e e e-9 9 e e e e e-9 69 e e-8 68 e e-9 79 e e-8 60 e-8 80 e e As shown n Table 3 wth the value of decreasng the standard devaton Std and the average number of ftness functon evaluatons FEs decrease as well whch means a better performance n optmzaton. From the statstc results shown n Table 3 t can be observed that 8 0 n general a settng of 0 ~ 0 preferably ft for most problems whch can be used as a common settng. As for low-dmensonal problems a settng of nearby the precson value leads to nce performances whle for a hgh-dmensonal problem the smaller s set the better result would be got. For example for Shfted 5 Sphere Functon wth 00-dmenson when 0 the FEs value decreases to 5983 better than when Analyss of Algorthm Convergence BGE s a type of randomzed optmzaton algorthms. The condtons of provng the convergence of a randomzed algorthm were frstly proposed by Sols and Wets [7]. They have gven the theorems to prove whether an algorthm has converged to the global optmal wth probablty whch can be summarzed as follows: Hypothess [7] f f ( ( z )) f ( z) S then f ( ( z )) f ( ). where s a functon to generate potental solutons s a random vector generated from the probablty space n (R B ) and f s the obectve functon. S whch s the subspace of R n represents the constrant space of the problem. s probablty measurement on B whch s the doman of R n subset. Hypothess [7] f A s a Borel subset of S satsfes ( A) 0 then ( ( A)) 0 0 where ( A) s a n -dmensonal closure of subset A and ( A) s a probablty ndcatng the rate that generates A. Theorem [7] Suppose f s a fathomable functon S s a fathomable subset of R n { z } 0 s a soluton sequence generated by the randomzed algorthm. If both Hypothess and Hypothess are satsfed smultaneously then lm P[ z ] where R represents the set of the global optmal solutons. Accordng to the theorem f both Hypothess and Hypothess are satsfed smultaneously for BGE t can be confrmed that the proposed BGE algorthm converges to the global optmal soluton wth probablty. The convergence proof of BGE s gven as follows: In the BGE algorthm the return value before the tth teraton s the functon value of x ( t ).e. f ( x ( t )) and f ( x ( t )) represents the functon value of the tth teraton value x () t where f( x ) represents R 9

10 J. L et al. the obectve functon. The functon of Hypothess s defned as: x ( t ) f f ( x ( t )) f ( x ( t)) ( x( t ) x( t)) x ( t) f f ( x ( t )) f ( x ( t)) It can be nferred Hypothess s satsfed for BGE. For Hypothess all that s needed s to prove the S - szed sample space contans S thus S M t S where M t represents the support set of the th ndvdual s sample space n tth teraton. Suppose there are N teratons n the search and the range of the th teraton s S whch s the support set as well. Therefore the unon space of the populaton (a set of ndvduals) s N S. The range of an ndvdual s adustable and when range covers the boundary of the soluton space though there are only a few ndvduals t can enable N S S Then Hypothess s satsfed for BGE algorthm. In concluson BGE converges to the global optmal soluton wth probablty accordng to the theorem. 6. Concluson In ths paper the BGE algorthm based on two crtcal models.e. bnary dfference and gravtatonal evoluton for global optmzaton was proposed. The proposed algorthm BGE was compared wth the fferental Evoluton by testng both algorthms on benchmar functons. Smulaton results show that the BGE can explore the soluton space more effectvely than E to obtan the global soluton and the BGE requres a much smaller sze populaton than E does. The parameters of BGE are studed and set by traland-error and the convergence analyss was also conducted to show BGE can converge to the global optmal soluton wth probablty. Certanly there s stll room to further mprove BGE. For nstance the number of parameters should be reduced to smplfy the algorthm and the clusterng and gravtatonal models should be studed to solve hgh-dmensonal problems whch are our future research wor. Acnowledgement Ths paper s partally supported by Natonal Natural Scence Foundaton of Chna under Grant Numbers ; and the Natural Scence Foundaton of Zheang Provnce under Grant Number Y References. J. H. Holland. Adaptaton n Natural and Artfcal Systems Unversty of Mchgan Press Ann Arbor MI Y. G. X T. Y. Cha and W. M. Yun Survey on Genetc Algorthm Control Theory and Applcatons vol.3 no.6 pp S. Krpatrc C.. Gelatt M. P. Vecch Optmzaton by Smulated Annealng Scence vol.0 no.598 pp K. Mor M. Tsuyama and T. Fuuda Immune Algorthm wth Searchng versty and ts Applcaton to Resource Allocaton Problem T.IEE Japan vol. 3-C no.0 pp M. J. L A. Luo and T. S. Tong Artfcal Immune Algorthm and Its Applcatons Control Theory and Applcatons vol. no L. Wang J. Pan and L. C. Jao The Immune Algorthm Acta Electronca Snca vol. 8 no R. C. Eberhart and Y. H. Sh Computatonal Intellgence: Concepts to Implementatons Elsever Sngapore J. Kennedy and R. C. Eberhart Partcle Swarm Optmzaton Proc. of the IEEE Int. Conf. on Neural Networs Pscataway NJ pp X. P. Zhang Y. P. u G. Q. Qn and Z. Qn Adaptve Partcle Swarm Algorthm wth ynamcally Changng Inerta Weght Journal of X an Jaotong Unversty vol. 39 no. 0 pp M. orgo and V. Manezzo A. Colorn. Ant System: Optmzaton by a Colony of Cooperatng Agents IEEE Transactons on Systems Man and Cybernetcs Part B vol.6 no. pp M. orgo M. Brattar and T. Stützle Ant Colony Optmzaton-Artfcal Ants as a Computatonal Intellgence Technque IEEE Computatonal Intellgence Magazne K. M. Passno Bacteral Foragng Optmzaton Internatonal Journal of Swarm Intellgence Research vol. no. pp V. Chrs E. Tsang Guded Local Search Techncal Report CSM-7 epartment of Computer Scence Unversty of Essex UK B. L. Webster Solvng Combnatoral Optmzaton Problems Usng a New Algorthm Based on Gravtatonal Attracton Ph.. thess Melbourne Florda Insttute of Technology S. R. Balachandar and K. Kannan Randomzed Gravtatonal Emulaton Search Algorthm for Symmetrc 9

11 Bnary fference Gravtatonal Evoluton Travelng Salesman Problem Appled Mathematcs and Computaton vol.9 pp Y. T. Hsao C. L. Chuang J. A. Jang and C. C. Chen A Novel Optmzaton Algorthm: Space Gravtatonal Optmzaton Proc. of 005 IEEE Internatonal Conference on Systems Man and Cybernetcs vol.3 pp C. L. Chuang and J. A. Jang Integrated Radaton Optmzaton: Inspred by the Gravtatonal Radaton n the Curvature of Space-tme Proc. of 007 IEEE Congress on Evolutonary Computaton pp E. Rashed Gravtatonal Search Algorthm Ph.. thess Shahd Bahonar Unversty of Kerman Kerman Iran R. Esmat N. P. Hossen and S. Saed GSA: A Gravtatonal Search Algorthm Informaton Scences vol.79 pp E. Rashed N. P. Hossen and S. Saed BGSA: Bnary Gravtatonal Search Algorthm. Natural Computng. vol.9 pp Q. Kang L. Wang and Q. Wu A Novel Self-organzng Partcle Swarm Optmzaton Based on Gravtaton Feld Model Proc. of the 007 Amercan Control Conference New Yor USA R. Storn and K. Pcce fferental Evoluton A Smple and Effcent Heurstc for Global Optmzaton over Contnuous Spaces Journal of Global Optmzaton vol. pp Q. W. Yang L. Ca and Y. C. Xue A Survey of fferental Evoluton Algorthms Pattern Recognton and Artfcal Intellgence vol. no. pp R. Mendes and A. S. Mohas yne: a fferental Evoluton for ynamc Optmzaton Problems Proc of the Congress on Evolutonary Optmzaton Ednburgh UK.vol.3 pp Y. Lou J. L. L and Y. S. Wang. A Bnary-fferental Evoluton Algorthm Based on Orderng of Indvduals Proc. of the Internatonal Conference on Natural Computaton. vol.5 pp Z. Y. Yang K. Tang and X. Yao Self-adaptve fferental Evoluton wth Neghborhood Search Proc. of the 008 IEEE Congress on Evolutonary Computaton pp F. J. Sols and R. T. Wets Mnmzaton by Random Search Technques. Mathematcs of Operatons Research vol.6 no. pp

The Study of Teaching-learning-based Optimization Algorithm

The Study of Teaching-learning-based Optimization Algorithm Advanced Scence and Technology Letters Vol. (AST 06), pp.05- http://dx.do.org/0.57/astl.06. The Study of Teachng-learnng-based Optmzaton Algorthm u Sun, Yan fu, Lele Kong, Haolang Q,, Helongang Insttute

More information

Markov Chain Monte Carlo Lecture 6

Markov Chain Monte Carlo Lecture 6 where (x 1,..., x N ) X N, N s called the populaton sze, f(x) f (x) for at least one {1, 2,..., N}, and those dfferent from f(x) are called the tral dstrbutons n terms of mportance samplng. Dfferent ways

More information

Using Immune Genetic Algorithm to Optimize BP Neural Network and Its Application Peng-fei LIU1,Qun-tai SHEN1 and Jun ZHI2,*

Using Immune Genetic Algorithm to Optimize BP Neural Network and Its Application Peng-fei LIU1,Qun-tai SHEN1 and Jun ZHI2,* Advances n Computer Scence Research (ACRS), volume 54 Internatonal Conference on Computer Networks and Communcaton Technology (CNCT206) Usng Immune Genetc Algorthm to Optmze BP Neural Network and Its Applcaton

More information

A Study on Improved Cockroach Swarm Optimization Algorithm

A Study on Improved Cockroach Swarm Optimization Algorithm A Study on Improved Cockroach Swarm Optmzaton Algorthm 1 epartment of Computer Scence and Engneerng, Huaan Vocatonal College of Informaton Technology,Huaan 223003, Chna College of Computer and Informaton,

More information

Power law and dimension of the maximum value for belief distribution with the max Deng entropy

Power law and dimension of the maximum value for belief distribution with the max Deng entropy Power law and dmenson of the maxmum value for belef dstrbuton wth the max Deng entropy Bngy Kang a, a College of Informaton Engneerng, Northwest A&F Unversty, Yanglng, Shaanx, 712100, Chna. Abstract Deng

More information

Wavelet chaotic neural networks and their application to continuous function optimization

Wavelet chaotic neural networks and their application to continuous function optimization Vol., No.3, 04-09 (009) do:0.436/ns.009.307 Natural Scence Wavelet chaotc neural networks and ther applcaton to contnuous functon optmzaton Ja-Ha Zhang, Yao-Qun Xu College of Electrcal and Automatc Engneerng,

More information

Lecture Notes on Linear Regression

Lecture Notes on Linear Regression Lecture Notes on Lnear Regresson Feng L fl@sdueducn Shandong Unversty, Chna Lnear Regresson Problem In regresson problem, we am at predct a contnuous target value gven an nput feature vector We assume

More information

A New Evolutionary Computation Based Approach for Learning Bayesian Network

A New Evolutionary Computation Based Approach for Learning Bayesian Network Avalable onlne at www.scencedrect.com Proceda Engneerng 15 (2011) 4026 4030 Advanced n Control Engneerng and Informaton Scence A New Evolutonary Computaton Based Approach for Learnng Bayesan Network Yungang

More information

Solving of Single-objective Problems based on a Modified Multiple-crossover Genetic Algorithm: Test Function Study

Solving of Single-objective Problems based on a Modified Multiple-crossover Genetic Algorithm: Test Function Study Internatonal Conference on Systems, Sgnal Processng and Electroncs Engneerng (ICSSEE'0 December 6-7, 0 Duba (UAE Solvng of Sngle-objectve Problems based on a Modfed Multple-crossover Genetc Algorthm: Test

More information

Differential Evolution Algorithm with a Modified Archiving-based Adaptive Tradeoff Model for Optimal Power Flow

Differential Evolution Algorithm with a Modified Archiving-based Adaptive Tradeoff Model for Optimal Power Flow 1 Dfferental Evoluton Algorthm wth a Modfed Archvng-based Adaptve Tradeoff Model for Optmal Power Flow 2 Outlne Search Engne Constrant Handlng Technque Test Cases and Statstcal Results 3 Roots of Dfferental

More information

Simulated Power of the Discrete Cramér-von Mises Goodness-of-Fit Tests

Simulated Power of the Discrete Cramér-von Mises Goodness-of-Fit Tests Smulated of the Cramér-von Mses Goodness-of-Ft Tests Steele, M., Chaselng, J. and 3 Hurst, C. School of Mathematcal and Physcal Scences, James Cook Unversty, Australan School of Envronmental Studes, Grffth

More information

A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS

A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS HCMC Unversty of Pedagogy Thong Nguyen Huu et al. A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS Thong Nguyen Huu and Hao Tran Van Department of mathematcs-nformaton,

More information

MODIFIED PARTICLE SWARM OPTIMIZATION FOR OPTIMIZATION PROBLEMS

MODIFIED PARTICLE SWARM OPTIMIZATION FOR OPTIMIZATION PROBLEMS Journal of Theoretcal and Appled Informaton Technology 3 st ecember 0. Vol. No. 005 0 JATIT & LLS. All rghts reserved. ISSN: 9985 www.jatt.org EISSN: 87395 MIFIE PARTICLE SARM PTIMIZATIN FR PTIMIZATIN

More information

Particle Swarm Optimization with Adaptive Mutation in Local Best of Particles

Particle Swarm Optimization with Adaptive Mutation in Local Best of Particles 1 Internatonal Congress on Informatcs, Envronment, Energy and Applcatons-IEEA 1 IPCSIT vol.38 (1) (1) IACSIT Press, Sngapore Partcle Swarm Optmzaton wth Adaptve Mutaton n Local Best of Partcles Nanda ulal

More information

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm Desgn and Optmzaton of Fuzzy Controller for Inverse Pendulum System Usng Genetc Algorthm H. Mehraban A. Ashoor Unversty of Tehran Unversty of Tehran h.mehraban@ece.ut.ac.r a.ashoor@ece.ut.ac.r Abstract:

More information

A destination swap scheme for multi-agent system with agent-robots in region search problems

A destination swap scheme for multi-agent system with agent-robots in region search problems A destnaton swap scheme for mult-agent system wth agent-robots n regon search problems Zongln Ye, Shuo Yang, Shuo Zhu, Yazhe Tang, Hu Cao, and Yanbn Zhang ABSTRACT As the mult-agent robots can share nformaton

More information

NUMERICAL DIFFERENTIATION

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

More information

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity Week3, Chapter 4 Moton n Two Dmensons Lecture Quz A partcle confned to moton along the x axs moves wth constant acceleraton from x =.0 m to x = 8.0 m durng a 1-s tme nterval. The velocty of the partcle

More information

A New Scrambling Evaluation Scheme based on Spatial Distribution Entropy and Centroid Difference of Bit-plane

A New Scrambling Evaluation Scheme based on Spatial Distribution Entropy and Centroid Difference of Bit-plane A New Scramblng Evaluaton Scheme based on Spatal Dstrbuton Entropy and Centrod Dfference of Bt-plane Lang Zhao *, Avshek Adhkar Kouch Sakura * * Graduate School of Informaton Scence and Electrcal Engneerng,

More information

Chapter 13: Multiple Regression

Chapter 13: Multiple Regression Chapter 13: Multple Regresson 13.1 Developng the multple-regresson Model The general model can be descrbed as: It smplfes for two ndependent varables: The sample ft parameter b 0, b 1, and b are used to

More information

The Minimum Universal Cost Flow in an Infeasible Flow Network

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

More information

Appendix B: Resampling Algorithms

Appendix B: Resampling Algorithms 407 Appendx B: Resamplng Algorthms A common problem of all partcle flters s the degeneracy of weghts, whch conssts of the unbounded ncrease of the varance of the mportance weghts ω [ ] of the partcles

More information

An Improved multiple fractal algorithm

An Improved multiple fractal algorithm Advanced Scence and Technology Letters Vol.31 (MulGraB 213), pp.184-188 http://dx.do.org/1.1427/astl.213.31.41 An Improved multple fractal algorthm Yun Ln, Xaochu Xu, Jnfeng Pang College of Informaton

More information

An improved multi-objective evolutionary algorithm based on point of reference

An improved multi-objective evolutionary algorithm based on point of reference IOP Conference Seres: Materals Scence and Engneerng PAPER OPEN ACCESS An mproved mult-objectve evolutonary algorthm based on pont of reference To cte ths artcle: Boy Zhang et al 08 IOP Conf. Ser.: Mater.

More information

Kernel Methods and SVMs Extension

Kernel Methods and SVMs Extension Kernel Methods and SVMs Extenson The purpose of ths document s to revew materal covered n Machne Learnng 1 Supervsed Learnng regardng support vector machnes (SVMs). Ths document also provdes a general

More information

An Adaptive Learning Particle Swarm Optimizer for Function Optimization

An Adaptive Learning Particle Swarm Optimizer for Function Optimization An Adaptve Learnng Partcle Swarm Optmzer for Functon Optmzaton Changhe L and Shengxang Yang Abstract Tradtonal partcle swarm optmzaton (PSO) suffers from the premature convergence problem, whch usually

More information

FUZZY GOAL PROGRAMMING VS ORDINARY FUZZY PROGRAMMING APPROACH FOR MULTI OBJECTIVE PROGRAMMING PROBLEM

FUZZY GOAL PROGRAMMING VS ORDINARY FUZZY PROGRAMMING APPROACH FOR MULTI OBJECTIVE PROGRAMMING PROBLEM Internatonal Conference on Ceramcs, Bkaner, Inda Internatonal Journal of Modern Physcs: Conference Seres Vol. 22 (2013) 757 761 World Scentfc Publshng Company DOI: 10.1142/S2010194513010982 FUZZY GOAL

More information

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud Resource Allocaton wth a Budget Constrant for Computng Independent Tasks n the Cloud Wemng Sh and Bo Hong School of Electrcal and Computer Engneerng Georga Insttute of Technology, USA 2nd IEEE Internatonal

More information

Beyond Zudilin s Conjectured q-analog of Schmidt s problem

Beyond Zudilin s Conjectured q-analog of Schmidt s problem Beyond Zudln s Conectured q-analog of Schmdt s problem Thotsaporn Ae Thanatpanonda thotsaporn@gmalcom Mathematcs Subect Classfcaton: 11B65 33B99 Abstract Usng the methodology of (rgorous expermental mathematcs

More information

Research on Route guidance of logistic scheduling problem under fuzzy time window

Research on Route guidance of logistic scheduling problem under fuzzy time window Advanced Scence and Technology Letters, pp.21-30 http://dx.do.org/10.14257/astl.2014.78.05 Research on Route gudance of logstc schedulng problem under fuzzy tme wndow Yuqang Chen 1, Janlan Guo 2 * Department

More information

The Convergence Speed of Single- And Multi-Objective Immune Algorithm Based Optimization Problems

The Convergence Speed of Single- And Multi-Objective Immune Algorithm Based Optimization Problems The Convergence Speed of Sngle- And Mult-Obectve Immune Algorthm Based Optmzaton Problems Mohammed Abo-Zahhad Faculty of Engneerng, Electrcal and Electroncs Engneerng Department, Assut Unversty, Assut,

More information

Comparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method

Comparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method Appled Mathematcal Scences, Vol. 7, 0, no. 47, 07-0 HIARI Ltd, www.m-hkar.com Comparson of the Populaton Varance Estmators of -Parameter Exponental Dstrbuton Based on Multple Crtera Decson Makng Method

More information

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

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

More information

Multi-Robot Formation Control Based on Leader-Follower Optimized by the IGA

Multi-Robot Formation Control Based on Leader-Follower Optimized by the IGA IOSR Journal of Computer Engneerng (IOSR-JCE e-issn: 2278-0661,p-ISSN: 2278-8727, Volume 19, Issue 1, Ver. III (Jan.-Feb. 2017, PP 08-13 www.osrjournals.org Mult-Robot Formaton Control Based on Leader-Follower

More information

The Second Anti-Mathima on Game Theory

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

More information

Durban Watson for Testing the Lack-of-Fit of Polynomial Regression Models without Replications

Durban Watson for Testing the Lack-of-Fit of Polynomial Regression Models without Replications Durban Watson for Testng the Lack-of-Ft of Polynomal Regresson Models wthout Replcatons Ruba A. Alyaf, Maha A. Omar, Abdullah A. Al-Shha ralyaf@ksu.edu.sa, maomar@ksu.edu.sa, aalshha@ksu.edu.sa Department

More information

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

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

More information

Comparative Analysis of SPSO and PSO to Optimal Power Flow Solutions

Comparative Analysis of SPSO and PSO to Optimal Power Flow Solutions Internatonal Journal for Research n Appled Scence & Engneerng Technology (IJRASET) Volume 6 Issue I, January 018- Avalable at www.jraset.com Comparatve Analyss of SPSO and PSO to Optmal Power Flow Solutons

More information

2016 Wiley. Study Session 2: Ethical and Professional Standards Application

2016 Wiley. Study Session 2: Ethical and Professional Standards Application 6 Wley Study Sesson : Ethcal and Professonal Standards Applcaton LESSON : CORRECTION ANALYSIS Readng 9: Correlaton and Regresson LOS 9a: Calculate and nterpret a sample covarance and a sample correlaton

More information

x = , so that calculated

x = , so that calculated Stat 4, secton Sngle Factor ANOVA notes by Tm Plachowsk n chapter 8 we conducted hypothess tests n whch we compared a sngle sample s mean or proporton to some hypotheszed value Chapter 9 expanded ths to

More information

Hongyi Miao, College of Science, Nanjing Forestry University, Nanjing ,China. (Received 20 June 2013, accepted 11 March 2014) I)ϕ (k)

Hongyi Miao, College of Science, Nanjing Forestry University, Nanjing ,China. (Received 20 June 2013, accepted 11 March 2014) I)ϕ (k) ISSN 1749-3889 (prnt), 1749-3897 (onlne) Internatonal Journal of Nonlnear Scence Vol.17(2014) No.2,pp.188-192 Modfed Block Jacob-Davdson Method for Solvng Large Sparse Egenproblems Hongy Mao, College of

More information

Operating conditions of a mine fan under conditions of variable resistance

Operating conditions of a mine fan under conditions of variable resistance Paper No. 11 ISMS 216 Operatng condtons of a mne fan under condtons of varable resstance Zhang Ynghua a, Chen L a, b, Huang Zhan a, *, Gao Yukun a a State Key Laboratory of Hgh-Effcent Mnng and Safety

More information

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

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

More information

APPLICATION OF RBF NEURAL NETWORK IMPROVED BY PSO ALGORITHM IN FAULT DIAGNOSIS

APPLICATION OF RBF NEURAL NETWORK IMPROVED BY PSO ALGORITHM IN FAULT DIAGNOSIS Journal of Theoretcal and Appled Informaton Technology 005-01 JATIT & LLS. All rghts reserved. ISSN: 199-8645 www.jatt.org E-ISSN: 1817-3195 APPLICATION OF RBF NEURAL NETWORK IMPROVED BY PSO ALGORITHM

More information

A Robust Method for Calculating the Correlation Coefficient

A Robust Method for Calculating the Correlation Coefficient A Robust Method for Calculatng the Correlaton Coeffcent E.B. Nven and C. V. Deutsch Relatonshps between prmary and secondary data are frequently quantfed usng the correlaton coeffcent; however, the tradtonal

More information

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

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

More information

An Extended Hybrid Genetic Algorithm for Exploring a Large Search Space

An Extended Hybrid Genetic Algorithm for Exploring a Large Search Space 2nd Internatonal Conference on Autonomous Robots and Agents Abstract An Extended Hybrd Genetc Algorthm for Explorng a Large Search Space Hong Zhang and Masum Ishkawa Graduate School of L.S.S.E., Kyushu

More information

The lower and upper bounds on Perron root of nonnegative irreducible matrices

The lower and upper bounds on Perron root of nonnegative irreducible matrices Journal of Computatonal Appled Mathematcs 217 (2008) 259 267 wwwelsevercom/locate/cam The lower upper bounds on Perron root of nonnegatve rreducble matrces Guang-Xn Huang a,, Feng Yn b,keguo a a College

More information

DETERMINATION OF UNCERTAINTY ASSOCIATED WITH QUANTIZATION ERRORS USING THE BAYESIAN APPROACH

DETERMINATION OF UNCERTAINTY ASSOCIATED WITH QUANTIZATION ERRORS USING THE BAYESIAN APPROACH Proceedngs, XVII IMEKO World Congress, June 7, 3, Dubrovn, Croata Proceedngs, XVII IMEKO World Congress, June 7, 3, Dubrovn, Croata TC XVII IMEKO World Congress Metrology n the 3rd Mllennum June 7, 3,

More information

4 Analysis of Variance (ANOVA) 5 ANOVA. 5.1 Introduction. 5.2 Fixed Effects ANOVA

4 Analysis of Variance (ANOVA) 5 ANOVA. 5.1 Introduction. 5.2 Fixed Effects ANOVA 4 Analyss of Varance (ANOVA) 5 ANOVA 51 Introducton ANOVA ANOVA s a way to estmate and test the means of multple populatons We wll start wth one-way ANOVA If the populatons ncluded n the study are selected

More information

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

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

More information

Economics 101. Lecture 4 - Equilibrium and Efficiency

Economics 101. Lecture 4 - Equilibrium and Efficiency Economcs 0 Lecture 4 - Equlbrum and Effcency Intro As dscussed n the prevous lecture, we wll now move from an envronment where we looed at consumers mang decsons n solaton to analyzng economes full of

More information

An Improved Particle Swarm Optimization Algorithm based on Membrane Structure

An Improved Particle Swarm Optimization Algorithm based on Membrane Structure IJCSI Internatonal Journal of Computer Scence Issues Vol. 10 Issue 1 No January 013 ISSN (Prnt): 1694-0784 ISSN (Onlne): 1694-0814 www.ijcsi.org 53 An Improved Partcle Swarm Optmzaton Algorthm based on

More information

Report on Image warping

Report on Image warping Report on Image warpng Xuan Ne, Dec. 20, 2004 Ths document summarzed the algorthms of our mage warpng soluton for further study, and there s a detaled descrpton about the mplementaton of these algorthms.

More information

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing Advanced Scence and Technology Letters, pp.164-168 http://dx.do.org/10.14257/astl.2013 Pop-Clc Nose Detecton Usng Inter-Frame Correlaton for Improved Portable Audtory Sensng Dong Yun Lee, Kwang Myung Jeon,

More information

VARIATION OF CONSTANT SUM CONSTRAINT FOR INTEGER MODEL WITH NON UNIFORM VARIABLES

VARIATION OF CONSTANT SUM CONSTRAINT FOR INTEGER MODEL WITH NON UNIFORM VARIABLES VARIATION OF CONSTANT SUM CONSTRAINT FOR INTEGER MODEL WITH NON UNIFORM VARIABLES BÂRZĂ, Slvu Faculty of Mathematcs-Informatcs Spru Haret Unversty barza_slvu@yahoo.com Abstract Ths paper wants to contnue

More information

One-sided finite-difference approximations suitable for use with Richardson extrapolation

One-sided finite-difference approximations suitable for use with Richardson extrapolation Journal of Computatonal Physcs 219 (2006) 13 20 Short note One-sded fnte-dfference approxmatons sutable for use wth Rchardson extrapolaton Kumar Rahul, S.N. Bhattacharyya * Department of Mechancal Engneerng,

More information

Solving Nonlinear Differential Equations by a Neural Network Method

Solving Nonlinear Differential Equations by a Neural Network Method Solvng Nonlnear Dfferental Equatons by a Neural Network Method Luce P. Aarts and Peter Van der Veer Delft Unversty of Technology, Faculty of Cvlengneerng and Geoscences, Secton of Cvlengneerng Informatcs,

More information

A Weighted Utility Framework for Mining Association Rules

A Weighted Utility Framework for Mining Association Rules A Weghted Utlty Framework for Mnng Assocaton Rules M Sulaman Khan 1, 2, Maybn Muyeba 1, Frans Coenen 2 1 School of Computng, Lverpool Hope Unversty, UK 2 Department of Computer Scence, Unversty of Lverpool,

More information

Société de Calcul Mathématique SA

Société de Calcul Mathématique SA Socété de Calcul Mathématque SA Outls d'ade à la décson Tools for decson help Probablstc Studes: Normalzng the Hstograms Bernard Beauzamy December, 202 I. General constructon of the hstogram Any probablstc

More information

Min Cut, Fast Cut, Polynomial Identities

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

More information

Evolutionary Computational Techniques to Solve Economic Load Dispatch Problem Considering Generator Operating Constraints

Evolutionary Computational Techniques to Solve Economic Load Dispatch Problem Considering Generator Operating Constraints Internatonal Journal of Engneerng Research and Applcatons (IJERA) ISSN: 48-96 Natonal Conference On Advances n Energy and Power Control Engneerng (AEPCE-K1) Evolutonary Computatonal Technques to Solve

More information

Week 5: Neural Networks

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

More information

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

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

More information

Chapter 2 Real-Coded Adaptive Range Genetic Algorithm

Chapter 2 Real-Coded Adaptive Range Genetic Algorithm Chapter Real-Coded Adaptve Range Genetc Algorthm.. Introducton Fndng a global optmum n the contnuous doman s challengng for Genetc Algorthms (GAs. Tradtonal GAs use the bnary representaton that evenly

More information

biologically-inspired computing lecture 21 Informatics luis rocha 2015 INDIANA UNIVERSITY biologically Inspired computing

biologically-inspired computing lecture 21 Informatics luis rocha 2015 INDIANA UNIVERSITY biologically Inspired computing lecture 21 -nspred Sectons I485/H400 course outlook Assgnments: 35% Students wll complete 4/5 assgnments based on algorthms presented n class Lab meets n I1 (West) 109 on Lab Wednesdays Lab 0 : January

More information

Assessment of Site Amplification Effect from Input Energy Spectra of Strong Ground Motion

Assessment of Site Amplification Effect from Input Energy Spectra of Strong Ground Motion Assessment of Ste Amplfcaton Effect from Input Energy Spectra of Strong Ground Moton M.S. Gong & L.L Xe Key Laboratory of Earthquake Engneerng and Engneerng Vbraton,Insttute of Engneerng Mechancs, CEA,

More information

V is the velocity of the i th

V is the velocity of the i th Proceedngs of the 007 IEEE Swarm Intellgence Symposum (SIS 007) Probablstcally rven Partcle Swarms for Optmzaton of Mult Valued screte Problems : esgn and Analyss Kalyan Veeramachanen, Lsa Osadcw, Ganapath

More information

Discrete Particle Swarm Optimization for TSP: Theoretical Results and Experimental Evaluations

Discrete Particle Swarm Optimization for TSP: Theoretical Results and Experimental Evaluations Dscrete Partcle Swarm Optmzaton for TSP: Theoretcal Results and Expermental Evaluatons Matthas Hoffmann, Mortz Mühlenthaler, Sabne Helwg, Rolf Wanka Department of Computer Scence, Unversty of Erlangen-Nuremberg,

More information

Annexes. EC.1. Cycle-base move illustration. EC.2. Problem Instances

Annexes. EC.1. Cycle-base move illustration. EC.2. Problem Instances ec Annexes Ths Annex frst llustrates a cycle-based move n the dynamc-block generaton tabu search. It then dsplays the characterstcs of the nstance sets, followed by detaled results of the parametercalbraton

More information

Assortment Optimization under MNL

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

More information

CHAPTER 14 GENERAL PERTURBATION THEORY

CHAPTER 14 GENERAL PERTURBATION THEORY CHAPTER 4 GENERAL PERTURBATION THEORY 4 Introducton A partcle n orbt around a pont mass or a sphercally symmetrc mass dstrbuton s movng n a gravtatonal potental of the form GM / r In ths potental t moves

More information

On the Multicriteria Integer Network Flow Problem

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

More information

On the correction of the h-index for career length

On the correction of the h-index for career length 1 On the correcton of the h-ndex for career length by L. Egghe Unverstet Hasselt (UHasselt), Campus Depenbeek, Agoralaan, B-3590 Depenbeek, Belgum 1 and Unverstet Antwerpen (UA), IBW, Stadscampus, Venusstraat

More information

CONTRAST ENHANCEMENT FOR MIMIMUM MEAN BRIGHTNESS ERROR FROM HISTOGRAM PARTITIONING INTRODUCTION

CONTRAST ENHANCEMENT FOR MIMIMUM MEAN BRIGHTNESS ERROR FROM HISTOGRAM PARTITIONING INTRODUCTION CONTRAST ENHANCEMENT FOR MIMIMUM MEAN BRIGHTNESS ERROR FROM HISTOGRAM PARTITIONING N. Phanthuna 1,2, F. Cheevasuvt 2 and S. Chtwong 2 1 Department of Electrcal Engneerng, Faculty of Engneerng Rajamangala

More information

Speeding up Computation of Scalar Multiplication in Elliptic Curve Cryptosystem

Speeding up Computation of Scalar Multiplication in Elliptic Curve Cryptosystem H.K. Pathak et. al. / (IJCSE) Internatonal Journal on Computer Scence and Engneerng Speedng up Computaton of Scalar Multplcaton n Ellptc Curve Cryptosystem H. K. Pathak Manju Sangh S.o.S n Computer scence

More information

CHAPTER-5 INFORMATION MEASURE OF FUZZY MATRIX AND FUZZY BINARY RELATION

CHAPTER-5 INFORMATION MEASURE OF FUZZY MATRIX AND FUZZY BINARY RELATION CAPTER- INFORMATION MEASURE OF FUZZY MATRI AN FUZZY BINARY RELATION Introducton The basc concept of the fuzz matr theor s ver smple and can be appled to socal and natural stuatons A branch of fuzz matr

More information

A Network Intrusion Detection Method Based on Improved K-means Algorithm

A Network Intrusion Detection Method Based on Improved K-means Algorithm Advanced Scence and Technology Letters, pp.429-433 http://dx.do.org/10.14257/astl.2014.53.89 A Network Intruson Detecton Method Based on Improved K-means Algorthm Meng Gao 1,1, Nhong Wang 1, 1 Informaton

More information

Discretization of Continuous Attributes in Rough Set Theory and Its Application*

Discretization of Continuous Attributes in Rough Set Theory and Its Application* Dscretzaton of Contnuous Attrbutes n Rough Set Theory and Its Applcaton* Gexang Zhang 1,2, Lazhao Hu 1, and Wedong Jn 2 1 Natonal EW Laboratory, Chengdu 610036 Schuan, Chna dylan7237@sna.com 2 School of

More information

The Order Relation and Trace Inequalities for. Hermitian Operators

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

More information

Semi-supervised Classification with Active Query Selection

Semi-supervised Classification with Active Query Selection Sem-supervsed Classfcaton wth Actve Query Selecton Jao Wang and Swe Luo School of Computer and Informaton Technology, Beng Jaotong Unversty, Beng 00044, Chna Wangjao088@63.com Abstract. Labeled samples

More information

A new construction of 3-separable matrices via an improved decoding of Macula s construction

A new construction of 3-separable matrices via an improved decoding of Macula s construction Dscrete Optmzaton 5 008 700 704 Contents lsts avalable at ScenceDrect Dscrete Optmzaton journal homepage: wwwelsevercom/locate/dsopt A new constructon of 3-separable matrces va an mproved decodng of Macula

More information

Global Sensitivity. Tuesday 20 th February, 2018

Global Sensitivity. Tuesday 20 th February, 2018 Global Senstvty Tuesday 2 th February, 28 ) Local Senstvty Most senstvty analyses [] are based on local estmates of senstvty, typcally by expandng the response n a Taylor seres about some specfc values

More information

Chapter Newton s Method

Chapter Newton s Method Chapter 9. Newton s Method After readng ths chapter, you should be able to:. Understand how Newton s method s dfferent from the Golden Secton Search method. Understand how Newton s method works 3. Solve

More information

on the improved Partial Least Squares regression

on the improved Partial Least Squares regression Internatonal Conference on Manufacturng Scence and Engneerng (ICMSE 05) Identfcaton of the multvarable outlers usng T eclpse chart based on the mproved Partal Least Squares regresson Lu Yunlan,a X Yanhu,b

More information

A Simple Inventory System

A Simple Inventory System A Smple Inventory System Lawrence M. Leems and Stephen K. Park, Dscrete-Event Smulaton: A Frst Course, Prentce Hall, 2006 Hu Chen Computer Scence Vrgna State Unversty Petersburg, Vrgna February 8, 2017

More information

Singular Value Decomposition: Theory and Applications

Singular Value Decomposition: Theory and Applications Sngular Value Decomposton: Theory and Applcatons Danel Khashab Sprng 2015 Last Update: March 2, 2015 1 Introducton A = UDV where columns of U and V are orthonormal and matrx D s dagonal wth postve real

More information

Design of College Discrete Mathematics Based on Particle Swarm Optimization. Jiedong Chen1, a

Design of College Discrete Mathematics Based on Particle Swarm Optimization. Jiedong Chen1, a 4th Internatonal Conference on Mechatroncs, Materals, Chemstry and Computer Engneerng (ICMMCCE 205) Desgn of College Dscrete Mathematcs Based on Partcle Swarm Optmzaton Jedong Chen, a Automotve Department,

More information

Some modelling aspects for the Matlab implementation of MMA

Some modelling aspects for the Matlab implementation of MMA Some modellng aspects for the Matlab mplementaton of MMA Krster Svanberg krlle@math.kth.se Optmzaton and Systems Theory Department of Mathematcs KTH, SE 10044 Stockholm September 2004 1. Consdered optmzaton

More information

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

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

More information

A DNA Coding Scheme for Searching Stable Solutions

A DNA Coding Scheme for Searching Stable Solutions A DNA odng Scheme for Searchng Stable Solutons Intaek Km, HeSong Lan, and Hwan Il Kang 2 Department of ommuncaton Eng., Myongj Unversty, 449-728, Yongn, South Korea kt@mju.ac.kr, hslan@hotmal.net 2 Department

More information

COS 521: Advanced Algorithms Game Theory and Linear Programming

COS 521: Advanced Algorithms Game Theory and Linear Programming COS 521: Advanced Algorthms Game Theory and Lnear Programmng Moses Charkar February 27, 2013 In these notes, we ntroduce some basc concepts n game theory and lnear programmng (LP). We show a connecton

More information

Chapter - 2. Distribution System Power Flow Analysis

Chapter - 2. Distribution System Power Flow Analysis Chapter - 2 Dstrbuton System Power Flow Analyss CHAPTER - 2 Radal Dstrbuton System Load Flow 2.1 Introducton Load flow s an mportant tool [66] for analyzng electrcal power system network performance. Load

More information

Markov Chain Monte Carlo (MCMC), Gibbs Sampling, Metropolis Algorithms, and Simulated Annealing Bioinformatics Course Supplement

Markov Chain Monte Carlo (MCMC), Gibbs Sampling, Metropolis Algorithms, and Simulated Annealing Bioinformatics Course Supplement Markov Chan Monte Carlo MCMC, Gbbs Samplng, Metropols Algorthms, and Smulated Annealng 2001 Bonformatcs Course Supplement SNU Bontellgence Lab http://bsnuackr/ Outlne! Markov Chan Monte Carlo MCMC! Metropols-Hastngs

More information

Grover s Algorithm + Quantum Zeno Effect + Vaidman

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

More information

VQ widely used in coding speech, image, and video

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

More information

Self-adaptive Differential Evolution Algorithm for Constrained Real-Parameter Optimization

Self-adaptive Differential Evolution Algorithm for Constrained Real-Parameter Optimization 26 IEEE Congress on Evolutonary Computaton Sheraton Vancouver Wall Centre Hotel, Vancouver, BC, Canada July 16-21, 26 Self-adaptve Dfferental Evoluton Algorthm for Constraned Real-Parameter Optmzaton V.

More information

Calculation of time complexity (3%)

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

More information

A Hybrid Algorithm Based on Gravitational Search and Particle Swarm Optimization Algorithm to Solve Function Optimization Problems

A Hybrid Algorithm Based on Gravitational Search and Particle Swarm Optimization Algorithm to Solve Function Optimization Problems Engneerng Letters, 25:1, EL_25_1_04 A Hybrd Algorthm Based on Gravtatonal Search and Partcle Swarm Optmzaton Algorthm to Solve Functon Optmzaton Problems Je-Sheng Wang, and Jang-D Song Abstract Gravtatonal

More information

Fuzzy Boundaries of Sample Selection Model

Fuzzy Boundaries of Sample Selection Model Proceedngs of the 9th WSES Internatonal Conference on ppled Mathematcs, Istanbul, Turkey, May 7-9, 006 (pp309-34) Fuzzy Boundares of Sample Selecton Model L. MUHMD SFIIH, NTON BDULBSH KMIL, M. T. BU OSMN

More information