Solving the Quadratic Eigenvalue Complementarity Problem by DC Programming

Size: px
Start display at page:

Download "Solving the Quadratic Eigenvalue Complementarity Problem by DC Programming"

Transcription

1 Solvng the Quadratc Egenvalue Complementarty Problem by DC Programmng Y-Shua Nu 1, Joaqum Júdce, Le Th Hoa An 3 and Pham Dnh Tao 4 1 Shangha JaoTong Unversty, Maths Departement and SJTU-Parstech, Chna nuyshua@sjtu.edu.cn Insttuto de Telecomuncações, Portugal judce@co.t.pt 3 Unversty of Lorrane, France hoa-an.le-th@unv-lorrane.fr 4 Natonal Insttute of Appled Scences, France pham@nsa-rouen.fr Abstract. We present n ths paper some results for solvng the Quadratc Egenvalue Complementarty Problem (QECP) by usng DC(Dfference of Convex functons) programmng approaches. Two equvalent Nonconvex Polynomal Programmng (NLP) formulatons of QECP are ntroduced. We focus on the constructon of the DC programmng formulatons of the QECP from these NLPs. The correspondng numercal soluton algorthms based on the classcal DC Algorthm (DCA) are also dscussed. Keywords: Egenvalue Problem, Complementarty Problem, Nonconvex Polynomal Programmng, DC Programmng, DCA. 1 Introducton Gven three matrces A, B, C R n n, the Quadratc Egenvalue Complementarty Problem (QECP) conssts of fndng a λ R and an assocated nonzero vector x R n such that w = λ Ax + λbx + Cx x T w = 0 x 0, w 0 (1) Ths problem and some applcatons have been frstly ntroduced n [19] and s usually denoted by QECP(A, B, C). In any soluton (λ, x) of QECP(A, B, C), the λ-component s called a quadratc complementary egenvalue, and the vector x-component s a quadratc complementary egenvector assocated to λ. QECP s an extenson of the well-known Egenvalue Complementarty Problem (ECP) [18], whch conssts of fndng a complementary egenvalue λ R

2 YS Nu et al. and an assocated complementary egenvector x R n \ {0} such that w = λbx Cx x T w = 0 x 0, w 0 where B, C R n n are two gven matrces. Clearly, ECP s a specal case of QECP where the matrx A s null. Durng the past several years, many applcatons of ECP have been dscussed and a number of algorthms have been proposed for the soluton of ths problem and some extensons [1 3, 6 11, 15, 16]. ECP has at least one soluton f the matrx B of the leadng λ-term s postve defnte (PD) [9, 18]. Contrary to the ECP, the QECP may have no soluton even when the matrx A of leadng λ-term s PD. For nstance, f B = 0, A, C are PD matrces, there s no soluton for QECP snce x T w = λ x T Ax+x T Cx > 0, λ R, x R n \ {0}. The exstence of a soluton to QECP depends on the gven (A, B, C). If the matrx A s PD, QECP has at least a soluton f one of the two followng condtons holds: () C / S 0 [4], where S 0 s the class of matrces defned by C S 0 x 0, x 0, Cx 0. () co-hyperbolcty[19] : (x T Bx) 4(x T Ax)(x T Cx) for all x 0, x 0. In practce, nvestgatng whether C S 0 reduces to solvng a specal lnear program [4]. On the other hand, t s relatvely hard to prove that cohyperbolcty holds. However, there are some suffcent condtons whch mply the co-hyperbolcty. For nstance, ths occurs f A and C are both PD matrces. A number of algorthms have been proposed for the soluton of QECP when A PD and one of the condtons C / S 0 or co-hyperbolcty holds [1, 4 6, 19]. As dscussed n [4 6], some of these methods are based on nonlnear programmng (NLP) formulatons of QECP such that (λ, x) s a soluton of QECP f and only f (λ, x) s a global mnmum of NLP wth an optmal value equal to zero. In ths paper, we ntroduce two nonlnear programmng formulatons and ther correspondng DC programmng formulatons when co-hyperbolcty holds, and we brefly dscuss the DC Algorthm for the soluton of these DC programs. The paper s organzed as follows. Secton contans the nonlnear programmng formulatons of QECP, and the correspondng dc formulatons mentoned before. A new result on lower and upper bounds estmaton of the quadratc complementary egenvalue s gven n secton 3. The numercal soluton algorthms for solvng these DC programmng formulatons are dscussed n secton 4. Some conclusons are presented n the last secton. DC Programmng Formulatons for QECP In ths secton, we ntroduce two DC programmng formulatons of QECP when A PD and the co-hyperbolc property holds. These DC programs are based ()

3 DCA for QECP 3 on two NLP formulatons of QECP. The constructon of the DC programmng problem requres lower and upper bounds on the λ-varable whch can be computed by the procedures dscussed n [6]. We wll also present a new procedure for such a goal n the secton 3..1 Nonlnear Programmng Formulatons As dscussed n [6], QECP s equvalent to the followng NLP: (P ) 0 = mn f(x, y, z, w, λ) := y λx + z λy + x T w s.t. w = Az + By + Cx e T x = 1 e T y = λ x 0, w 0, z 0. (3) As (x, y, z, w, λ) s an optmal soluton of the problem (P ) f and only f (λ, x) s a soluton of QECP. In fact, for any soluton of QECP (λ, x) that does not satsfy e T x x = 1, we can always construct a soluton (λ, ) of QECP satsfyng e T x such a constrant. The problem (P ) s a polynomal programmng problem where a nonconvex polynomal functon f(x, y, z, w, λ) s mnmzed subject to lnear constrants. Due to the fact that any polynomal functon s a dc functon, we can reformulate the problem (P ) as a dc program. On the other hand, observng that the complementarty constrant w T x = 0, x 0, w 0 holds f and only f w T x = n =1 mn(x, w ) = 0, we have the followng equvalent nonlnear programmng formulaton of (P ): (P ) 0 = mn f (x, y, z, w, λ) = y λx + z λy + n =1 mn(x, w ) s.t w = Az + By + Cx e T x = 1 e T y = λ x 0, w 0, z 0. The problems (P ) and (P ) have the same set of lnear constrants. The dffculty for solvng (P ) and (P ) reles on the non-convexty on ther objectve functons.. DC programmng formulatons The polynomal functon f n (P ) can be decomposed nto four parts: f(x, y, z, w, λ) = y + z λy T (x + z) + λ ( x + y ) + x T w = f 0 (y, z) + f 1 (x, y, z, λ) + f (x, y, λ) + f 3 (x, w) wth f 0 (y, z) = y + z f 1 (x, y, z, λ) = λy T (x + z) f (x, y, λ) = λ ( x + y ) f 3 (x, w) = x T w

4 4 YS Nu et al. The functon f 0 s convex quadratc functon, whle f 1, f, f 3 are nonconvex polynomal functons. Smlarly, the objectve functon f n (P ) s also decomposed nto the followng four terms as: f (x, y, z, w, λ) = f 0 (y, z) + f 1 (x, y, z, λ) + f (x, y, λ) + f 3 (x, w) where f 3 (x, w) defned by n =1 mn(x, w ) s a polyhedral concave functon. Both the blnear functon f 3 and the polyhedral concave functon f 3 are classcal dc functons whose dc decompostons are as follows: 1. DC decomposton of blnear functon f 3 : f 3 (x, w) = x + w 4 x w 4 n whch x+w 4 and x w 4 are both convex quadratc functons.. DC decomposton of polyhedral functon f 3 : (4) f 3 (x, w) = n n mn(x, w ) = (0) ( mn(x, w )) (5) =1 =1 where n =1 mn(x, w ) s a convex polyhedral functon. To obtan a dc decompostons of the nonconvex polynomal functons f 1 and f, we frst obtan the expressons of ther gradents and hessans: 1. Gradent and Hessan of f 1 : x f 1 (x, y, z, λ) f 1 (x, y, z, λ) = y f 1 (x, y, z, λ) z f 1 (x, y, z, λ) = λ f 1 (x, y, z, λ) λy λ(x + z) λy y T (x + z). 0 λi 0 y f 1 (x, y, z, λ) = λi 0 λi (x + z) 0 λi 0 y. y T (x + z) T y T 0. Gradent and Hessan of f : x f (x, y, λ) λ x f (x, y, λ) = y f (x, y, λ) = λ y. λ f (x, y, λ) λ( x + y ) λ I 0 4λx f (x, y, z, λ) = 0 λ I 4λy 4λx T 4λy T ( x + y )

5 DCA for QECP 5 The spectral radus of the hessan matrces f 1 and f (denoted by ρ( f 1 ) and ρ( f )) can be bounded above by the nduced 1-norm as follows: ρ( f 1 ) f 1 1 = max{ λ + y, x + z + λ, ( y + x + z )} ρ( f ) f 1 = max{λ + λ x, λ + λ y, x + y + λ ( x + y )} Thus ρ( f 1 ) and ρ( f )) are bounded when the varables (x, y, z, w, λ) of (P ) and (P ) are bounded. The next proposton shows that f the quadratc complementary egenvalue λ of QECP s bounded, then the varables x, y, z, w are bounded wth respect to the bounds of λ. Proposton 1. If the quadratc complementary egenvalue λ of QECP s bounded n an nterval [l, u], then any optmal soluton of (P ) and (P ) satsfes: x [0, 1] n ; y [mn{0, l}, max{0, u}] n ; z [0, max{u, l }] n ; max{u, l } j A 1j + max{ l, u } j B 1j + j C 1j 0 w. max{u, l } j A nj + max{ l, u } j B nj +. j C nj Proof. Suppose that we could determne some values l and u such that λ- component of QECP s located n the nterval [l, u]. 1. e T x = 1, x 0 mples x [0, 1] n.. y = λx, x [0, 1] n and λ [l, u] mply y [mn{0, l}, max{0, u}] n. 3. z = λy, y = λx z = λ x, wth x [0, 1] n, λ [l, u], leads to z [0, max{u, l }] n. 4. Snce w 0, the upper bound of w s obtaned from the defnton of w as Az+By+Cx. As x [0, 1] n, y [mn{0, l}, max{0, u}] n, z [0, max{u, l }] n, then w s also bounded: w max{u, l } j A 1j + max{ l, u } j B 1j + j C 1j. max{u, l } j A nj + max{ l, u } j B nj + j C nj Let us defne the convex polyhedral set: C := {(x, y, z, w, λ) : w = Az + By + Cx, e T x = 1, e T y = λ, x [0, 1] n,. y [mn{0, l}, max{0, u}] n, z [0, max{u, l }] n, w 0, l λ u}. The problems (P ) and (P ) defned on C have the same set of optmal solutons, and ρ( f 1 ) and ρ( f ) are bounded. In fact, the followng proposton holds:

6 6 YS Nu et al. Proposton. For (x, y, z, w, λ) C, ρ( f 1 ) + n(p + p) = ρ 1 ρ( f ) (3np + p + 1) = ρ where p = max{ l, u }. Proof. Snce λ [l, u], then λ max{ l, u } = p. Hence, ρ( f 1 ) max{ λ + y, x + z + λ, ( y + x + z )}. But, Hence, x + z y np. x + z 1 + np. ρ( f 1 ) max{p, 1 + p + p, 1 + n(p + p)} = + n(p + p) = ρ 1 Smlarly, ρ( f ) max{λ + λ x, λ + λ y, x + y + λ ( x + y )} max{p + p, 3p, 3np + p + 1} = (3np + p + 1) = ρ. Thus, we get a dc decomposton for f 1 and f as follows: f 1 (x, y, z, λ) = ρ 1 (x, y, z, λ) ( ρ 1 (x, y, z, λ) f 1 (x, y, z, λ)) f (x, y, λ) = ρ (x, y, λ) ( ρ (x, y, λ) f (x, y, λ)) where ρ1 (x, y, z, λ) and ρ (x, y, λ) are quadratc convex functons. Whle ρ 1 (x, y, z, λ) f 1 (x, y, z, λ) and ρ (x, y, λ) f (x, y, λ) are locally convex restrcted on C. Usng the dc decompostons of f 1,f,f 3 and f 3 derved n ths secton, we get the followng dc decomposton for the objectve functons f and f. 1. A dc decomposton for f s gven by: where g(x, y, z, w, λ) = f(x, y, z, w, λ) = g(x, y, z, w, λ) h(x, y, z, w, λ) x + w 4 + ρ 1 + ρ x +( ρ 1 + ρ +1) y +( ρ 1 h(x, y, z, w, λ) = g(x, y, z, w, λ) f(x, y, z, w, λ). +1) z + ρ 1 + ρ λ,

7 DCA for QECP 7. A dc decomposton for f s gven by: where f (x, y, z, w, λ) = g (x, y, z, w, λ) h (x, y, z, w, λ) g (x, y, z, λ) = ρ 1 + ρ x + ( ρ 1 + ρ + 1) y + ( ρ 1 + 1) z + ρ 1 + ρ λ, h (x, y, z, w, λ) = g (x, y, z, λ) f (x, y, z, w, λ). The functons g and g are both convex quadratc functons, whle h and h are locally convex functons restrcted on the convex polyhedral set C. Fnally, we get the followng equvalent DC programs of (P ) and (P ) as below: (P DC ) 0 = mn g(x, y, z, w, λ) h(x, y, z, w, λ) (6) s.t. (x, y, z, w, λ) C. (P DC ) 0 = mn g (x, y, z, λ) h (x, y, z, w, λ) s.t. (x, y, z, w, λ) C. (7) 3 Lower and upper bounds for the quadratc complementary egenvalue λ Snce the bounds of the varables x, y, z, w n C, as well as the dc decompostons gven n the prevous secton depend on the bounds of λ, we need to estmate ts upper and lower bounds. The followng theorem gves these values. Proposton 3. If A PD and the co-hyperbolc condton holds, the λ-component of any soluton of QECP satsfes l = β α λ γ + α = u wth s = mn{x T Ax : e T x = 1, x 0}, α = max{γ, β } + max,j{ Cj} β = γ = { mn{ Bj} max{a, f mn{ B j} j} > 0; mn{ B j} s, f mn{ B j } 0. { max{ Bj} s, f max{ B j } > 0; max{ B j} max{a j}, f max{ B j} 0. s, Proof. Snce A PD and the co-hyperbolc condton holds, the λ-component of any soluton of QECP satsfes λ = xt Bx ± (x T Bx) 4(x T Ax)(x T Cx) x T. Ax

8 8 YS Nu et al. Let U = {e T x = 1, x 0}. For a gven matrx M R n n and for any x U, we next prove that: mn,j M j x T Mx max M j, x U. (8),j If fact, let (Mx) denote the -th element of the vector Mx. Then But, (Mx) s bounded by j=1 x T Mx = n x (Mx) =1 n n mn{ M j x j : x U} (Mx) max{ M j x j : x U}, x U. Snce the lnear programs mn{ n j=1 M jx j : x U} and max{ n j=1 M jx j : x U} have optmal solutons on vertces, the optmal values of the above lnear programs are exactly mn j {M j } and max j {M j }. Hence, we can compute bounds for x T Mx on U as follows: j=1 mn,j {M j} = mn{ x mn{m j } : x U} j x mn{m j } j x (Mx) = x T Mx x max{m j } max{ j x max j {M j } : x U} = max,j {M,j}. Hence, (8) s true. Usng the bounds (8) for the matrces B and C, we have: Snce A PD, we have Accordngly, xt Bx x T Ax mn{ B j } x T Ax mn { B j} x T Bx max { B j},,j,j mn { C j} x T Cx max,j,j { C j} 0 < s = mn{x T Ax : x U} x T Ax max,j {A j}, x U. s bounded by: xt Bx x T Ax max{ B j} x T γ = Ax { max{ Bj} s, f max{ B j } > 0; max{ B j} max{a j}, f max{ B j} 0. and mn{ B j } x T Ax β = { mn{ Bj} max{a j}, f mn{ B j} > 0; mn{ B j} s, f mn{ B j } 0.

9 DCA for QECP 9 Then ( xt Bx x T Ax ) + xt Cx x T Ax max{γ, β } + max{ C j} = α. s Fnally, we can compute bounds for λ as follows: β α xt Bx x T Ax ( xt Bx x T Ax ) + xt Cx x T Ax λ xt Bx x T Ax + ( xt Bx x T Ax ) + xt Cx x T Ax γ + α. In practce, t s nterestng to compare n the future the bound proposed here wth the one gven n [6]. The bounds gven n ths paper have been desgned such that they can be computed n a small amount of effort, even for large-scale problems. 4 DC Algorthms for solvng P DC and P DC In ths secton, we nvestgate how to solve the DC programmng formulatons (P DC ) and (P DC ). Gven a general DC program: mn{g(x) h(x) : x C}, where C s a non-empty convex set, the general DC algorthm (DCA) conssts of constructng two sequences {x k } and {y k } va the followng scheme[1 14]: x k y k h(x k ) x k+1 g (y k ) = argmn{g(x) x, y k : x C}. The symbol h stands for the sub-dfferental of the convex functon h, and g s the conjugate functon of g. These defntons are fundamental and can be found n any textbook of the convex analyss (see for example [17]). The sequence {x k } and {y k } are respectvely canddates for optmal solutons of the prmal and dual DC programs. In DCA, two major computatons should be consdered: 1. Computng h(x k ) to get y k.. Solvng the convex program argmn{g(x) x, y k : x C} to obtan x k+1. Now, we nvestgate the use of DCA to solve the DC programs (P DC ) and (P DC ). Concernng to (P DC ), snce the functon h s dfferentable, h(x, y, z, w, λ) s

10 10 YS Nu et al. reduced to a sngleton { h(x, y, z, w, λ)}, where h(x, y, z, w, λ) = g(x, y, z, w, λ) f(x, y, z, w, λ) x+w + (ρ 1 + ρ )x + λy λ x w (ρ 1 + ρ λ )y + λ(x + z) = ρ 1 z + λy w x. (9) (ρ 1 + ρ ( x + y ))λ + y T (x + z) For (P DC ), snce the functon h s non-dfferentable, we compute the convex set h (x, y, z, w, λ) as follows: (ρ 1 + ρ )x + λy λ x u (ρ 1 + ρ λ )y + λ(x + z) h (x, y, z, w, λ) = ρ 1 z + λy (10) v (ρ 1 + ρ ( x + y ))λ + y T (x + z) where 1, x < w ; u = (u ) =1,...,n, u = {0, 1}, x = w ; 0, x > w. 0, x < w ; v = (v ) =1,...,n, v = {0, 1}, x = w ; 1, x > w. Fnally, DCA appled to (P DC ) and (P DC ) requres solvng respectvely one convex quadratc program over a polyhedral convex set n each teraton. The followng two fxed-pont schemes descrbe our dc algorthms: (x k+1, y k+1, z k+1, w k+1, λ k+1 ) = argmn{g(x, y, z, w, λ) (x, y, z, w, λ), h(x k, y k, z k, w k, λ k ) : (x, y, z, w, λ) C} (11) wth g(x, y, z, w, λ) = x+w ρ 1+ρ λ. 4 + ρ1+ρ x + ( ρ1+ρ + 1) y + ( ρ1 + 1) z + (x k+1, y k+1, z k+1, w k+1, λ k+1 ) = argmn{g (x, y, z, λ) (x, y, z, w, λ), Y k : (x, y, z, w, λ) C} (1) wth Y k h (x k, y k, z k, w k, λ k ) and g (x, y, z, λ) = ρ1+ρ x + ( ρ1+ρ 1) y + ( ρ1 + 1) z + ρ1+ρ λ. + These convex quadratc programs can be effcently solved va a quadratc programmng solver such as CPLEX, Gurob, XPress, etc. DCA should termnate f one of the followng stoppng crtera s satsfed for gven tolerances ɛ 1, ɛ and ɛ 3.

11 DCA for QECP 11 (1) The sequence {(x k, y k, z k, w k, λ k )} converges,.e., (x k+1, y k+1, z k+1, w k+1, λ k+1 ) (x k, y k, z k, w k, λ k ) ɛ 1 () The sequence {f(x k, y k, z k, w k, λ k )} (resp. {f (x k, y k, z k, w k, λ k )}) converges,.e., f(x k+1, y k+1, z k+1, w k+1, λ k+1 ) f(x k, y k, z k, w k, λ k ) ɛ (resp. f (x k+1, y k+1, z k+1, w k+1, λ k+1 ) f (x k, y k, z k, w k, λ k ) ɛ ). (3) The suffcent global ɛ-optmalty condton holds,.e., f(x k, y k, z k, w k, λ k ) ɛ 3 (resp. f (x k, y k, z k, w k, λ k ) ɛ 3 ). The followng theorem ndcates the convergence of DCA: Theorem 1 (Convergence theorem of DCA). DCA appled to QECP generates convergence sequences {(x k, y k, z k, w k, λ k )} and {f(x k, y k, z k, w k, λ k )} (resp. {f (x k, y k, z k, w k, λ k )}) such that: The sequence {f(x k, y k, z k, w k, λ k )} (resp. {f (x k, y k, z k, w k, λ k )}) s decreasng and bounded below. The sequence {(x k, y k, z k, w k, λ k )} converges ether to a soluton of QECP when the thrd stoppng condton s satsfed or to a general KKT pont of (P DC ) (resp. (P DC )). Proof. The proof of the theorem s an obvous consequence of the general convergence theorem of DCA [1 14]. The suffcent global optmalty condton s due to the fact that the optmal value of the dc program s equal to zero. 5 Conclusons In ths paper, we have presented two DC programmng formulatons of the Quadratc Egenvalue Complementarty Problem. The correspondng numercal soluton algorthms based on the classcal DCA for solvng these dc programs were brefly dscussed. The numercal results and the analyss of the performance of DCA for solvng QECP wll be gven n a future paper. We wll dscuss a new local dc decomposton algorthm that s desgned to speed up the convergence of DCA. Furthermore, that paper wll also be devoted to the soluton of QECP when the condton A PD and C / S 0 holds. A new DC formulaton of QECP based on the reformulaton of an equvalent extended ECP wll be ntroduced to deal wth ths case and the correspondng DC Algorthm wll be dscussed.

12 1 YS Nu et al. Acknowledgements The research of Y-Shua Nu n ths project s partally supported and fnanced by the Innovatve Research Fund of Shangha Jao Tong Unversty 985 Program. Joaqum Júdce was partally supported n the scope of R&D Unt UID/EEA/5008/013, fnanced by the applcable fnancal framework (FCT/MEC through natonal funds and the applcable co-funded by FEDER-PT00 partnershp agreement). References 1. Adly, S., Seeger, A. A non-smooth algorthm for cone constraned egenvalue problems. Computatonal Optmzaton and Applcatons. Vol. 49, (011).. Adly, S., Rammal, H. A new method for solvng second-order cone egenvalue complementarty problem. Journal of Optmzaton Theory and Applcatons. (014), do: /s Brás, C, Fukushma, M.. Júdce, J., Rosa, S. Varatonal nequalty formulaton for the asymmetrc egenvalue complementarty problem and ts soluton by means of a gap functon. Pacfc Journal of Optmzaton. Vol. 8, (01). 4. Brás, C, Iusem, A.N.. Júdce, J. On the quadratc egenvalue complementarty problem. To appear n Journal of Global Optmzaton. 5. Fernandes, L. M., Júdce, J., Fukushma, M., Iusem, A. On the symmetrc quadratc egenvalue complementarty problem. Optmzaton Methods and Software, Vol. 9, (014). 6. Fernandes, L.M., Júdce, J., Sheral, H.D., Forjaz M.A., On an enumeratve algorthm for solvng egenvalue complementarty problems. Computatonal Optmzaton and Applcatons. Vol. 59, (014). 7. Júdce, J, Sheral, H. D., Rbero, I. The egenvalue complementarty problem. Computatonal Optmzaton and Applcatons. Vol. 37, (007). 8. Júdce, J., Raydan, M., Rosa, S., Santos, S. On the soluton of the symmetrc complementarty problem by the spectral projected gradent method. Numercal Algorthms. Vol. 44, (008). 9. Júdce, J., Sheral, H. D., Rbero, I., Rosa, S. On the asymmetrc egenvalue complementarty problem. Optmzaton Methods and Software. Vol. 4, (009). 10. Le Th, H.A., Moen, M., Pham, D.T., Judce, J. A DC programmng approach for solvng the symmetrc Egenvalue Complementarty Problem, Computatonal Optmzaton and Applcatons, Vol. 51, (01). 11. Nu, Y.S., Le Th, H.A., Pham, D.T., Júdce, J., Effcent dc programmng approaches for the asymmetrc egenvalue complementarty problem. Optmzaton Methods and Software. Vol. 8, (013). 1. Pham Dnh, T., Le Th, H.A.: DC optmzaton algorthms for solvng the trust regon subproblem, SIAM Journal of Optmzaton, Vol. 8, (1998). 13. Pham Dnh, T., Le Th, H.A.: DC Programmng. Theory, Algorthms, Applcatons: The State of the Art. Frst Internatonal Whorkshop on Global Constraned Optmzaton and Constrant Satsfacton, Nce, October 4 (00). 14. Pham Dnh, T., Le Th, H.A.: The DC programmng and DCA Revsted wth DC Models of Real World Nonconvex Optmzaton Problems. Annals of Operatons Research, Vol. 133, 3 46 (005).

13 DCA for QECP Pnto da Costa, A., Seeger, A. Cone constraned egenvalue problems, theory and algorthms. Computatonal Optmzaton and Applcatons. Vol. 45, 5 57 (010). 16. Queroz, M.. Júdce, J.. Humes, C. The symmetrc egenvalue complementarty problem. Mathematcs of Computaton. Vol. 73, (003). 17. Rockafellar, R. Tyrell. Convex Analyss. Prnceton: Prnceton Unversty Press (1970). 18. Seeger, A. Egenvalue analyss of equlbrum processes defned by lnear complementartv condtons. Lnear Algebra and Its Applcatons. Vol. 94, 1 14 (1999). 19. Seeger, A. Quadratc egenvalue problems under conc constrants. SIAM Journal on Matrx Analyss and Applcatons. Vol. 3, (011).

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

MMA and GCMMA two methods for nonlinear optimization

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

More information

Yong Joon Ryang. 1. Introduction Consider the multicommodity transportation problem with convex quadratic cost function. 1 2 (x x0 ) T Q(x x 0 )

Yong Joon Ryang. 1. Introduction Consider the multicommodity transportation problem with convex quadratic cost function. 1 2 (x x0 ) T Q(x x 0 ) Kangweon-Kyungk Math. Jour. 4 1996), No. 1, pp. 7 16 AN ITERATIVE ROW-ACTION METHOD FOR MULTICOMMODITY TRANSPORTATION PROBLEMS Yong Joon Ryang Abstract. The optmzaton problems wth quadratc constrants often

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

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

Perron Vectors of an Irreducible Nonnegative Interval Matrix

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

More information

On a direct solver for linear least squares problems

On a direct solver for linear least squares problems ISSN 2066-6594 Ann. Acad. Rom. Sc. Ser. Math. Appl. Vol. 8, No. 2/2016 On a drect solver for lnear least squares problems Constantn Popa Abstract The Null Space (NS) algorthm s a drect solver for lnear

More information

Solutions to exam in SF1811 Optimization, Jan 14, 2015

Solutions to exam in SF1811 Optimization, Jan 14, 2015 Solutons to exam n SF8 Optmzaton, Jan 4, 25 3 3 O------O -4 \ / \ / The network: \/ where all lnks go from left to rght. /\ / \ / \ 6 O------O -5 2 4.(a) Let x = ( x 3, x 4, x 23, x 24 ) T, where the varable

More information

Inexact Newton Methods for Inverse Eigenvalue Problems

Inexact Newton Methods for Inverse Eigenvalue Problems Inexact Newton Methods for Inverse Egenvalue Problems Zheng-jan Ba Abstract In ths paper, we survey some of the latest development n usng nexact Newton-lke methods for solvng nverse egenvalue problems.

More information

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

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

More information

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017 U.C. Berkeley CS94: Beyond Worst-Case Analyss Handout 4s Luca Trevsan September 5, 07 Summary of Lecture 4 In whch we ntroduce semdefnte programmng and apply t to Max Cut. Semdefnte Programmng Recall that

More information

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

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

More information

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

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

Perfect Competition and the Nash Bargaining Solution

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

More information

6.854J / J Advanced Algorithms Fall 2008

6.854J / J Advanced Algorithms Fall 2008 MIT OpenCourseWare http://ocw.mt.edu 6.854J / 18.415J Advanced Algorthms Fall 2008 For nformaton about ctng these materals or our Terms of Use, vst: http://ocw.mt.edu/terms. 18.415/6.854 Advanced Algorthms

More information

A Local Variational Problem of Second Order for a Class of Optimal Control Problems with Nonsmooth Objective Function

A Local Variational Problem of Second Order for a Class of Optimal Control Problems with Nonsmooth Objective Function A Local Varatonal Problem of Second Order for a Class of Optmal Control Problems wth Nonsmooth Objectve Functon Alexander P. Afanasev Insttute for Informaton Transmsson Problems, Russan Academy of Scences,

More information

SELECTED SOLUTIONS, SECTION (Weak duality) Prove that the primal and dual values p and d defined by equations (4.3.2) and (4.3.3) satisfy p d.

SELECTED SOLUTIONS, SECTION (Weak duality) Prove that the primal and dual values p and d defined by equations (4.3.2) and (4.3.3) satisfy p d. SELECTED SOLUTIONS, SECTION 4.3 1. Weak dualty Prove that the prmal and dual values p and d defned by equatons 4.3. and 4.3.3 satsfy p d. We consder an optmzaton problem of the form The Lagrangan for ths

More information

10-801: Advanced Optimization and Randomized Methods Lecture 2: Convex functions (Jan 15, 2014)

10-801: Advanced Optimization and Randomized Methods Lecture 2: Convex functions (Jan 15, 2014) 0-80: Advanced Optmzaton and Randomzed Methods Lecture : Convex functons (Jan 5, 04) Lecturer: Suvrt Sra Addr: Carnege Mellon Unversty, Sprng 04 Scrbes: Avnava Dubey, Ahmed Hefny Dsclamer: These notes

More information

Solutions HW #2. minimize. Ax = b. Give the dual problem, and make the implicit equality constraints explicit. Solution.

Solutions HW #2. minimize. Ax = b. Give the dual problem, and make the implicit equality constraints explicit. Solution. Solutons HW #2 Dual of general LP. Fnd the dual functon of the LP mnmze subject to c T x Gx h Ax = b. Gve the dual problem, and make the mplct equalty constrants explct. Soluton. 1. The Lagrangan s L(x,

More information

APPENDIX A Some Linear Algebra

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

More information

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

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

More information

STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2018 LECTURE 16

STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2018 LECTURE 16 STAT 39: MATHEMATICAL COMPUTATIONS I FALL 218 LECTURE 16 1 why teratve methods f we have a lnear system Ax = b where A s very, very large but s ether sparse or structured (eg, banded, Toepltz, banded plus

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

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

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

ρ some λ THE INVERSE POWER METHOD (or INVERSE ITERATION) , for , or (more usually) to

ρ some λ THE INVERSE POWER METHOD (or INVERSE ITERATION) , for , or (more usually) to THE INVERSE POWER METHOD (or INVERSE ITERATION) -- applcaton of the Power method to A some fxed constant ρ (whch s called a shft), x λ ρ If the egenpars of A are { ( λ, x ) } ( ), or (more usually) to,

More information

More metrics on cartesian products

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

More information

Convexity preserving interpolation by splines of arbitrary degree

Convexity preserving interpolation by splines of arbitrary degree Computer Scence Journal of Moldova, vol.18, no.1(52), 2010 Convexty preservng nterpolaton by splnes of arbtrary degree Igor Verlan Abstract In the present paper an algorthm of C 2 nterpolaton of dscrete

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

A Hybrid Variational Iteration Method for Blasius Equation

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

More information

Lecture 12: Discrete Laplacian

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

More information

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

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

More information

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

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

e - c o m p a n i o n

e - c o m p a n i o n OPERATIONS RESEARCH http://dxdoorg/0287/opre007ec e - c o m p a n o n ONLY AVAILABLE IN ELECTRONIC FORM 202 INFORMS Electronc Companon Generalzed Quantty Competton for Multple Products and Loss of Effcency

More information

Research Article Global Sufficient Optimality Conditions for a Special Cubic Minimization Problem

Research Article Global Sufficient Optimality Conditions for a Special Cubic Minimization Problem Mathematcal Problems n Engneerng Volume 2012, Artcle ID 871741, 16 pages do:10.1155/2012/871741 Research Artcle Global Suffcent Optmalty Condtons for a Specal Cubc Mnmzaton Problem Xaome Zhang, 1 Yanjun

More information

Norms, Condition Numbers, Eigenvalues and Eigenvectors

Norms, Condition Numbers, Eigenvalues and Eigenvectors Norms, Condton Numbers, Egenvalues and Egenvectors 1 Norms A norm s a measure of the sze of a matrx or a vector For vectors the common norms are: N a 2 = ( x 2 1/2 the Eucldean Norm (1a b 1 = =1 N x (1b

More information

6) Derivatives, gradients and Hessian matrices

6) Derivatives, gradients and Hessian matrices 30C00300 Mathematcal Methods for Economsts (6 cr) 6) Dervatves, gradents and Hessan matrces Smon & Blume chapters: 14, 15 Sldes by: Tmo Kuosmanen 1 Outlne Defnton of dervatve functon Dervatve notatons

More information

Stanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011

Stanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011 Stanford Unversty CS359G: Graph Parttonng and Expanders Handout 4 Luca Trevsan January 3, 0 Lecture 4 In whch we prove the dffcult drecton of Cheeger s nequalty. As n the past lectures, consder an undrected

More information

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

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

More information

LOW BIAS INTEGRATED PATH ESTIMATORS. James M. Calvin

LOW BIAS INTEGRATED PATH ESTIMATORS. James M. Calvin Proceedngs of the 007 Wnter Smulaton Conference S G Henderson, B Bller, M-H Hseh, J Shortle, J D Tew, and R R Barton, eds LOW BIAS INTEGRATED PATH ESTIMATORS James M Calvn Department of Computer Scence

More information

( ) 2 ( ) ( ) Problem Set 4 Suggested Solutions. Problem 1

( ) 2 ( ) ( ) Problem Set 4 Suggested Solutions. Problem 1 Problem Set 4 Suggested Solutons Problem (A) The market demand functon s the soluton to the followng utlty-maxmzaton roblem (UMP): The Lagrangean: ( x, x, x ) = + max U x, x, x x x x st.. x + x + x y x,

More information

Introduction. - The Second Lyapunov Method. - The First Lyapunov Method

Introduction. - The Second Lyapunov Method. - The First Lyapunov Method Stablty Analyss A. Khak Sedgh Control Systems Group Faculty of Electrcal and Computer Engneerng K. N. Toos Unversty of Technology February 2009 1 Introducton Stablty s the most promnent characterstc of

More information

Some Comments on Accelerating Convergence of Iterative Sequences Using Direct Inversion of the Iterative Subspace (DIIS)

Some Comments on Accelerating Convergence of Iterative Sequences Using Direct Inversion of the Iterative Subspace (DIIS) Some Comments on Acceleratng Convergence of Iteratve Sequences Usng Drect Inverson of the Iteratve Subspace (DIIS) C. Davd Sherrll School of Chemstry and Bochemstry Georga Insttute of Technology May 1998

More information

Approximate D-optimal designs of experiments on the convex hull of a finite set of information matrices

Approximate D-optimal designs of experiments on the convex hull of a finite set of information matrices Approxmate D-optmal desgns of experments on the convex hull of a fnte set of nformaton matrces Radoslav Harman, Mára Trnovská Department of Appled Mathematcs and Statstcs Faculty of Mathematcs, Physcs

More information

Consistency & Convergence

Consistency & Convergence /9/007 CHE 374 Computatonal Methods n Engneerng Ordnary Dfferental Equatons Consstency, Convergence, Stablty, Stffness and Adaptve and Implct Methods ODE s n MATLAB, etc Consstency & Convergence Consstency

More information

General viscosity iterative method for a sequence of quasi-nonexpansive mappings

General viscosity iterative method for a sequence of quasi-nonexpansive mappings Avalable onlne at www.tjnsa.com J. Nonlnear Sc. Appl. 9 (2016), 5672 5682 Research Artcle General vscosty teratve method for a sequence of quas-nonexpansve mappngs Cuje Zhang, Ynan Wang College of Scence,

More information

Proseminar Optimierung II. Victor A. Kovtunenko SS 2012/2013: LV

Proseminar Optimierung II. Victor A. Kovtunenko SS 2012/2013: LV Prosemnar Optmerung II Vctor A. Kovtunenko Insttute for Mathematcs and Scentfc Computng, Karl-Franzens Unversty of Graz, Henrchstr. 36, 8010 Graz, Austra; Lavrent ev Insttute of Hydrodynamcs, Sberan Dvson

More information

On the Global Linear Convergence of the ADMM with Multi-Block Variables

On the Global Linear Convergence of the ADMM with Multi-Block Variables On the Global Lnear Convergence of the ADMM wth Mult-Block Varables Tany Ln Shqan Ma Shuzhong Zhang May 31, 01 Abstract The alternatng drecton method of multplers ADMM has been wdely used for solvng structured

More information

Convex Optimization. Optimality conditions. (EE227BT: UC Berkeley) Lecture 9 (Optimality; Conic duality) 9/25/14. Laurent El Ghaoui.

Convex Optimization. Optimality conditions. (EE227BT: UC Berkeley) Lecture 9 (Optimality; Conic duality) 9/25/14. Laurent El Ghaoui. Convex Optmzaton (EE227BT: UC Berkeley) Lecture 9 (Optmalty; Conc dualty) 9/25/14 Laurent El Ghaou Organsatonal Mdterm: 10/7/14 (1.5 hours, n class, double-sded cheat sheet allowed) Project: Intal proposal

More information

Computing Correlated Equilibria in Multi-Player Games

Computing Correlated Equilibria in Multi-Player Games Computng Correlated Equlbra n Mult-Player Games Chrstos H. Papadmtrou Presented by Zhanxang Huang December 7th, 2005 1 The Author Dr. Chrstos H. Papadmtrou CS professor at UC Berkley (taught at Harvard,

More information

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

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

More information

Control of Uncertain Bilinear Systems using Linear Controllers: Stability Region Estimation and Controller Design

Control of Uncertain Bilinear Systems using Linear Controllers: Stability Region Estimation and Controller Design Control of Uncertan Blnear Systems usng Lnear Controllers: Stablty Regon Estmaton Controller Desgn Shoudong Huang Department of Engneerng Australan Natonal Unversty Canberra, ACT 2, Australa shoudong.huang@anu.edu.au

More information

The Finite Element Method: A Short Introduction

The Finite Element Method: A Short Introduction Te Fnte Element Metod: A Sort ntroducton Wat s FEM? Te Fnte Element Metod (FEM) ntroduced by engneers n late 50 s and 60 s s a numercal tecnque for solvng problems wc are descrbed by Ordnary Dfferental

More information

Interactive Bi-Level Multi-Objective Integer. Non-linear Programming Problem

Interactive Bi-Level Multi-Objective Integer. Non-linear Programming Problem Appled Mathematcal Scences Vol 5 0 no 65 3 33 Interactve B-Level Mult-Objectve Integer Non-lnear Programmng Problem O E Emam Department of Informaton Systems aculty of Computer Scence and nformaton Helwan

More information

Lecture 10 Support Vector Machines II

Lecture 10 Support Vector Machines II Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed

More information

COMPUTING THE NORM OF A MATRIX

COMPUTING THE NORM OF A MATRIX COMPUTING THE NORM OF A MATRIX KEITH CONRAD 1. Introducton In R n there s a standard noton of length: the sze of a vector v = (a 1,..., a n ) s v = a 2 1 + + a2 n. We wll dscuss n Secton 2 the general

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

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis A Appendx for Causal Interacton n Factoral Experments: Applcaton to Conjont Analyss Mathematcal Appendx: Proofs of Theorems A. Lemmas Below, we descrbe all the lemmas, whch are used to prove the man theorems

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

Global Optimization of Truss. Structure Design INFORMS J. N. Hooker. Tallys Yunes. Slide 1

Global Optimization of Truss. Structure Design INFORMS J. N. Hooker. Tallys Yunes. Slide 1 Slde 1 Global Optmzaton of Truss Structure Desgn J. N. Hooker Tallys Yunes INFORMS 2010 Truss Structure Desgn Select sze of each bar (possbly zero) to support the load whle mnmzng weght. Bar szes are dscrete.

More information

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

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

More information

A property of the elementary symmetric functions

A property of the elementary symmetric functions Calcolo manuscrpt No. (wll be nserted by the edtor) A property of the elementary symmetrc functons A. Esnberg, G. Fedele Dp. Elettronca Informatca e Sstemstca, Unverstà degl Stud della Calabra, 87036,

More information

8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS

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

More information

Lecture 3. Ax x i a i. i i

Lecture 3. Ax x i a i. i i 18.409 The Behavor of Algorthms n Practce 2/14/2 Lecturer: Dan Spelman Lecture 3 Scrbe: Arvnd Sankar 1 Largest sngular value In order to bound the condton number, we need an upper bound on the largest

More information

Errors for Linear Systems

Errors for Linear Systems Errors for Lnear Systems When we solve a lnear system Ax b we often do not know A and b exactly, but have only approxmatons  and ˆb avalable. Then the best thng we can do s to solve ˆx ˆb exactly whch

More information

Single-Facility Scheduling over Long Time Horizons by Logic-based Benders Decomposition

Single-Facility Scheduling over Long Time Horizons by Logic-based Benders Decomposition Sngle-Faclty Schedulng over Long Tme Horzons by Logc-based Benders Decomposton Elvn Coban and J. N. Hooker Tepper School of Busness, Carnege Mellon Unversty ecoban@andrew.cmu.edu, john@hooker.tepper.cmu.edu

More information

Problem Set 9 Solutions

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

More information

NP-Completeness : Proofs

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

More information

On the convergence of the block nonlinear Gauss Seidel method under convex constraints

On the convergence of the block nonlinear Gauss Seidel method under convex constraints Operatons Research Letters 26 (2000) 127 136 www.elsever.com/locate/orms On the convergence of the bloc nonlnear Gauss Sedel method under convex constrants L. Grppo a, M. Scandrone b; a Dpartmento d Informatca

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

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

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

Randić Energy and Randić Estrada Index of a Graph

Randić Energy and Randić Estrada Index of a Graph EUROPEAN JOURNAL OF PURE AND APPLIED MATHEMATICS Vol. 5, No., 202, 88-96 ISSN 307-5543 www.ejpam.com SPECIAL ISSUE FOR THE INTERNATIONAL CONFERENCE ON APPLIED ANALYSIS AND ALGEBRA 29 JUNE -02JULY 20, ISTANBUL

More information

Vector Norms. Chapter 7 Iterative Techniques in Matrix Algebra. Cauchy-Bunyakovsky-Schwarz Inequality for Sums. Distances. Convergence.

Vector Norms. Chapter 7 Iterative Techniques in Matrix Algebra. Cauchy-Bunyakovsky-Schwarz Inequality for Sums. Distances. Convergence. Vector Norms Chapter 7 Iteratve Technques n Matrx Algebra Per-Olof Persson persson@berkeley.edu Department of Mathematcs Unversty of Calforna, Berkeley Math 128B Numercal Analyss Defnton A vector norm

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

Lecture 10: May 6, 2013

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

More information

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

Randomized block proximal damped Newton method for composite self-concordant minimization

Randomized block proximal damped Newton method for composite self-concordant minimization Randomzed block proxmal damped Newton method for composte self-concordant mnmzaton Zhaosong Lu June 30, 2016 Revsed: March 28, 2017 Abstract In ths paper we consder the composte self-concordant CSC mnmzaton

More information

Appendix B. The Finite Difference Scheme

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

More information

Another converse of Jensen s inequality

Another converse of Jensen s inequality Another converse of Jensen s nequalty Slavko Smc Abstract. We gve the best possble global bounds for a form of dscrete Jensen s nequalty. By some examples ts frutfulness s shown. 1. Introducton Throughout

More information

An efficient algorithm for multivariate Maclaurin Newton transformation

An efficient algorithm for multivariate Maclaurin Newton transformation Annales UMCS Informatca AI VIII, 2 2008) 5 14 DOI: 10.2478/v10065-008-0020-6 An effcent algorthm for multvarate Maclaurn Newton transformaton Joanna Kapusta Insttute of Mathematcs and Computer Scence,

More information

Inner Product. Euclidean Space. Orthonormal Basis. Orthogonal

Inner Product. Euclidean Space. Orthonormal Basis. Orthogonal Inner Product Defnton 1 () A Eucldean space s a fnte-dmensonal vector space over the reals R, wth an nner product,. Defnton 2 (Inner Product) An nner product, on a real vector space X s a symmetrc, blnear,

More information

Deriving the X-Z Identity from Auxiliary Space Method

Deriving the X-Z Identity from Auxiliary Space Method Dervng the X-Z Identty from Auxlary Space Method Long Chen Department of Mathematcs, Unversty of Calforna at Irvne, Irvne, CA 92697 chenlong@math.uc.edu 1 Iteratve Methods In ths paper we dscuss teratve

More information

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

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

More information

MATH Homework #2

MATH Homework #2 MATH609-601 Homework #2 September 27, 2012 1. Problems Ths contans a set of possble solutons to all problems of HW-2. Be vglant snce typos are possble (and nevtable). (1) Problem 1 (20 pts) For a matrx

More information

Numerical Heat and Mass Transfer

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

More information

Lecture 21: Numerical methods for pricing American type derivatives

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

More information

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

EEE 241: Linear Systems

EEE 241: Linear Systems EEE : Lnear Systems Summary #: Backpropagaton BACKPROPAGATION The perceptron rule as well as the Wdrow Hoff learnng were desgned to tran sngle layer networks. They suffer from the same dsadvantage: they

More information

System of implicit nonconvex variationl inequality problems: A projection method approach

System of implicit nonconvex variationl inequality problems: A projection method approach Avalable onlne at www.tjnsa.com J. Nonlnear Sc. Appl. 6 (203), 70 80 Research Artcle System of mplct nonconvex varatonl nequalty problems: A projecton method approach K.R. Kazm a,, N. Ahmad b, S.H. Rzv

More information

1 Derivation of Point-to-Plane Minimization

1 Derivation of Point-to-Plane Minimization 1 Dervaton of Pont-to-Plane Mnmzaton Consder the Chen-Medon (pont-to-plane) framework for ICP. Assume we have a collecton of ponts (p, q ) wth normals n. We want to determne the optmal rotaton and translaton

More information

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

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

More information

Linear Approximation with Regularization and Moving Least Squares

Linear Approximation with Regularization and Moving Least Squares Lnear Approxmaton wth Regularzaton and Movng Least Squares Igor Grešovn May 007 Revson 4.6 (Revson : March 004). 5 4 3 0.5 3 3.5 4 Contents: Lnear Fttng...4. Weghted Least Squares n Functon Approxmaton...

More information

Difference Equations

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

More information

Maximizing the number of nonnegative subsets

Maximizing the number of nonnegative subsets Maxmzng the number of nonnegatve subsets Noga Alon Hao Huang December 1, 213 Abstract Gven a set of n real numbers, f the sum of elements of every subset of sze larger than k s negatve, what s the maxmum

More information

APPROXIMATE PRICES OF BASKET AND ASIAN OPTIONS DUPONT OLIVIER. Premia 14

APPROXIMATE PRICES OF BASKET AND ASIAN OPTIONS DUPONT OLIVIER. Premia 14 APPROXIMAE PRICES OF BASKE AND ASIAN OPIONS DUPON OLIVIER Prema 14 Contents Introducton 1 1. Framewor 1 1.1. Baset optons 1.. Asan optons. Computng the prce 3. Lower bound 3.1. Closed formula for the prce

More information

PHYS 705: Classical Mechanics. Calculus of Variations II

PHYS 705: Classical Mechanics. Calculus of Variations II 1 PHYS 705: Classcal Mechancs Calculus of Varatons II 2 Calculus of Varatons: Generalzaton (no constrant yet) Suppose now that F depends on several dependent varables : We need to fnd such that has a statonary

More information

STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2018 LECTURE 17. a ij x (k) b i. a ij x (k+1) (D + L)x (k+1) = b Ux (k)

STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2018 LECTURE 17. a ij x (k) b i. a ij x (k+1) (D + L)x (k+1) = b Ux (k) STAT 309: MATHEMATICAL COMPUTATIONS I FALL 08 LECTURE 7. sor method remnder: n coordnatewse form, Jacob method s = [ b a x (k) a and Gauss Sedel method s = [ b a = = remnder: n matrx form, Jacob method

More information