Pith. sign in

REVIEW 3 major objections 4 minor 45 references

An Efficient and Globally Optimal Algorithm for Nonconvex QCQP with One Equality Constraint

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper claims that the nonconvex single-equality-constraint QCQP is globally solvable in $O(N^3)$ time by simultaneous diagonalization followed by a bisection search on the KKT equation.

desk verdict A clean generic-case method whose universal global-optimality claim breaks in the trust-region hard case; reject as is, but fixable. read the letter →

arxiv 2506.03692 v1 pith:O3CVQFZN submitted 2025-06-04 math.OC

Licheng Zhao , Rui Zhou , Wenqiang Pu This is my paper
classification math.OC MSC 90C2690C2290C2065K05
keywords nonconvexQCQPsingleequalityconstraintglobaloptimalitysimultaneousdiagonalizationbisectionsearchKKTconditionssemidefiniterelaxationmatrix-format
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tackles a nonconvex quadratically constrained quadratic program (QCQP) with exactly one equality constraint: minimize a quadratic objective subject to a fixed second-order level set. In the standard case where the constraint matrix is positive definite ($A_1\succ 0$), it claims the global optimum can be found in guaranteed $O(N^3)$ time by a two-stage procedure. First, simultaneous diagonalization turns the two quadratic forms into a decoupled diagonal problem; second, the first- and second-order KKT conditions reduce the search for the optimal Lagrange multiplier to solving a one-dimensional monotone equation $f(\lambda)=c$ by bisection. The paper also extends the scheme to rank-deficient constraint matrices, indefinite full-rank constraints with a strictly convex objective, extra linear equalities, and matrix-format complex variables. If right, this replaces semidefinite relaxation (SDR) as the standard subproblem solver in applications from portfolio design to hybrid precoding, with far better speed and scalability.

What carries the argument

The load-bearing object is the pair: simultaneous diagonalization plus the scalar KKT equation. Simultaneous diagonalization finds an invertible $T$ with $A_1=T T^T$ and $A_0=T\operatorname{diag}(a)T^T$, and the affine map $y=T^T x+T^{-1}b_1$ converts problem (1) into $\min \sum_i (a_i y_i^2-2b_i y_i)$ subject to $\sum_i y_i^2=c$. The bisection stage then solves $f(\lambda)=\sum_i b_i^2/(a_i+\lambda)^2=c$ on the interval $(-\min_i a_i,+\infty)$, where the second-order KKT condition forces $\lambda\ge -\min_i a_i$; monotonicity of $f$ makes the root unique, so a simple bisection recovers the globally optimal $y^\star=b/(a+\lambda^\star\mathbf{1})$ and hence $x^\star$.

What would settle it

Set $N=2$, $A_0=\operatorname{diag}(0,1)$, $A_1=I$, $b_0=b_1=0$, $c_0=0$, $c_1=-1$. The constraint becomes $y_1^2+y_2^2=1$ and the objective is $y_2^2$, so the global minimum is $0$, attained at $(\pm1,0)$. Yet $c=1$ and $f(\lambda)=0$ for every $\lambda$, so the bisection stage of Algorithm 1 has no root to find and returns no solution. This instance isolates exactly the missing nonzero-$b$ assumption and would refute the claim that a root exists for every feasible instance.

Watch

Extended reading notes

Core claim

The central claim is that problem (1) has a unique KKT point that is globally optimal, and that this point can be recovered from the solution of a scalar equation. With $A_1\succ 0$, the paper proves that for any feasible instance there is an affine map $y=T^T x+T^{-1}b_1$ making the constraint $y^T y=c$ and the objective diagonal; then the first-order condition gives $y_i=b_i/(a_i+\lambda)$, and the second-order condition restricts $\lambda$ to $(-\min_i a_i,+\infty)$. In that interval $f(\lambda)=\sum_i b_i^2/(a_i+\lambda)^2$ is strictly decreasing, so $f(\lambda)=c$ has a unique root $\lambda^\star$, and the inverse map applied to $y^\star=b/(a+\lambda^\star\mathbf{1})$ is the global minimizer. The paper further claims that the same two-stage logic, with a modified $f$, handles rank-deficient $A_1$, indefinite $A_1$ under $A_0\succ 0$, additional linear equalities, and complex matrix variables, and reports numerics showing constraint violations below $10^{-5}$, optimality gaps below $10^{-4}$, and speeds two or more orders of magnitude above SDR.

Load-bearing premise

The proof assumes that every index attaining the smallest diagonal entry $a_i$ has a nonzero coefficient $b_i$, so that $f(\lambda)\to+\infty$ at the left endpoint and $f(\lambda)=c$ has a root for every $c>0$.

Editorial extensions

If this is right

  • For any feasible single-equality-constraint QCQP with $A_1\succ 0$, global optimality can be certified in $O(N^3)$ time without solving a semidefinite program, making the subproblem affordable at dimension up to $10^3$.
  • The same two-stage logic, with modified $f(\lambda)$, covers rank-deficient $A_1$ with a null-space elimination condition, indefinite full-rank $A_1$ when $A_0\succ 0$, and additional linear equalities via a null-space parametrization.
  • For complex matrix-format variables, the paper's phase-alignment argument reduces the multiplier equation to the column-norm quantities $b_j=\sqrt{\sum_i |b_{i,j}|^2}$, lowering complexity from $O((N_1N_2)^{6.5})$ to $O(N_1^3)$.
  • Numerical experiments in the paper report constraint violations below $10^{-5}$, optimality gaps below $10^{-4}$, runtime at least two orders of magnitude shorter than the SDR benchmark, and scalability to problem sizes at least ten times larger.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A practical implementation would need a separate fallback for the hard case in which $b_i=0$ at an index attaining $\min_i a_i$; the paper does not provide that routine, since its proof of the root's existence relies on $b_i\neq 0$ there.
  • The column-norm reduction in the matrix extension suggests a general principle: whenever the constraint is a Frobenius-norm equality and the two quadratic forms are simultaneously diagonalizable, only the column norms of the transformed variable enter the multiplier equation, so the effective dimension is the row size, not the full matrix size.
  • Because single-equality-constraint QCQPs appear inside iterative majorization and block-coordinate outer loops, a fast exact subproblem solver could accelerate those outer algorithms; this follows from the paper's complexity claim but is not spelled out.
  • A direct stress test would be to run SD-BS on deliberately constructed hard-case instances (zero coefficients at the minimal $a_i$) and compare against a trust-region solver that handles endpoint solutions; the paper's numerical section does not appear to include such instances.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper proposes SD-BS, a two-stage algorithm for the nonconvex QCQP with one equality constraint (Eq. (1)). In the standard case A1≻0, the SD stage simultaneously diagonalizes A0 and A1 through an affine change of variables, reducing the problem to the decoupled form (10). The BS stage then finds the Lagrange multiplier λ by solving the equation f(λ)=c, where f is defined in Eq. (15), using monotonicity on the interval (−min_i a_i, +∞). The paper claims that this procedure yields a globally optimal solution in O(N^3) time and extends the approach to rank-deficient A1, indefinite A1, additional linear constraints, and matrix-format variables. Numerical comparisons against SDR and fmincon are reported for randomized instances.

Significance. If the central claim were correct, the SD-BS algorithm would be a valuable contribution, offering O(N^3) complexity and a non-iterative (up to bisection) solution for a problem class typically solved by SDR with much higher complexity. The derivations are clean for generic data, and the reported experiments show excellent performance on randomly generated instances. However, the universal global-optimality guarantee for the standard case is false, because the proof of the endpoint behavior of f(λ) fails when the coefficient b_i vanishes for every index attaining the minimal a_i. In that trust-region hard case the global minimizer lies at λ=−min a_i with nonzero components in the minimal block, which Algorithm 1 cannot find. The numerical experiments generate generic data and therefore do not expose this failure. The contribution is thus significant only under a restrictive assumption that the paper does not state.

major comments (3)
  1. [Section II-B, Eq. (15) and the paragraph after Eq. (15)] The assertion that f(λ) → +∞ as λ → (−min_i a_i)^+ is false when b_i = 0 for every index i attaining the minimum of the a_i. In that case the limit is the finite value Σ_{a_i > min a} b_i^2/(a_i − min a)^2, so for c larger than this limit the equation f(λ)=c has no root in (−min_i a_i, +∞). For example, the standard-case instance min x^2 subject to x^2 = 2 (A0=A1=1, b0=b1=0, c1=−2) yields a=1, b=0, c=2 and f(λ)=0 for all λ; a global minimizer exists (any x=±√2) but no λ in (−1, +∞) satisfies f(λ)=2. This directly disproves the claimed unique-root existence and the universal global-optimality guarantee of Algorithm 1 for the standard case.
  2. [Section II-B, Eqs. (13) and (18)] The KKT solution formula y_i = b_i/(a_i + λ) excludes λ = −min_i a_i because the denominator would vanish. This is not merely a technical edge case: when the minimal-a_i block has b_i=0, the KKT system has solutions at λ=−min_i a_i with arbitrary y_i on that block, constrained only by the norm equation (11), and these solutions can be the global minimizers. Since Algorithm 1 searches only λ ∈ (−min_i a_i, +∞) via bisection, it cannot recover such solutions. Consequently, the claim in Section II-B that the first- and second-order KKT system has a unique KKT point is incorrect in general.
  3. [Section VII-A, experiment generation] The randomized instances in Section VII-A generate b0 and b1 with i.i.d. continuous Gaussian entries, so with probability one every b_i is nonzero. The failure mode described above, in which all b_i at the minimal a_i vanish, is never exercised. The numerical results are therefore consistent with the algorithm's behavior only in the generic case and provide no evidence for the universal claim made in the abstract and in Section II-B.
minor comments (4)
  1. [Abstract and Section II-C] The abstract calls the algorithm 'non-iterative', but the Bisection Search stage in Algorithm 1 is an iterative procedure; the complexity discussion in Section II-C should state the bisection iteration count or clarify that 'non-iterative' refers to the absence of iterative convex-optimization solves.
  2. [Introduction, first paragraph] The phrase 'type pf optimization problem' contains a typo and should read 'type of optimization problem'.
  3. [Appendix C, last paragraph] The sentence 'Plugging (40) into (40)' should refer to plugging Eq. (61) into Eq. (60); the current reference is self-referential and unclear.
  4. [Theorem 4, Eq. (41)] The formula arg(y_i) = arg(b_i) is undefined when b_i = 0; the theorem requires a convention or an assumption excluding zero b_i, which is another instance of the zero-coefficient case that the paper does not address.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the SD-BS derivation is a self-contained KKT construction, with no fitted input renamed as prediction and no load-bearing self-citation.

full rationale

SD-BS is a constructive algorithm: Section II-A transforms (1) into (10) via the invertible affine map of Theorem 2, and Section II-B derives the KKT stationarity equation f(λ)=c and the recovery map y_i=b_i/(a_i+λ). The global-optimality claim rests on existence of the root and the argument that the global minimizer must be among the unique second-order KKT point; this is a mathematical derivation from first principles, not a parameter fit, and no predicted quantity is defined in terms of fitted data. The only self-citations ([23], [40]) are contextual application/reference citations and do not carry the derivation. The proof has a genuine soundness gap at Eqs. (15)-(18): the asserted endpoint limit f(λ)→+∞ as λ→(−min a_i)^+ fails when all b_i at the minimal a_i vanish (the trust-region hard case), so the covered instances are only the generic ones where the minimizer is interior; this is a correctness limitation, not circularity. Accordingly no circular step is exhibited.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

No free parameters are fitted; the proof uses standard simultaneous diagonalization and KKT conditions. The main unstated axiom is the non-degeneracy condition (extremal b_i nonzero) that guarantees a secular equation root; this is what breaks in the hard case.

assumptions (3)
  • standard math For symmetric A0 and A1 ≻ 0, there exists invertible T with A0 = T Λ0 T^T and A1 = T T^T (Lemma 1).
    Standard simultaneous diagonalization via Cholesky factor and EVD; cited to [44].
  • domain assumption A global minimizer of problem (10) must satisfy the first-order KKT conditions, and the unique KKT point satisfying the second-order condition is globally optimal.
    The paper asserts this in Section II-B: 'the optimal solution must lie among the KKT points' and 'this system turns out to have a unique KKT point, so it automatically generates the globally optimal solution'. No sufficiency proof is given for this nonconvex equality-constrained problem; it is only valid in the generic case without the hard case.
  • ad hoc to paper For the index i with a_i = min_j a_j, the coefficient b_i is nonzero, so f(λ) → +∞ as λ → (-min a_i)^+.
    Used implicitly in Section II-B to guarantee a unique root of f(λ)=c for every c>0. It is not stated as an assumption and is false when the extremal b_i is zero, producing the hard case.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An Efficient and Globally Optimal Algorithm for Nonconvex QCQP with One Equality Constraint." pith.science (2026). https://pith.science/paper/O3CVQFZN

@misc{pith2026250603692,
  author       = {Pith},
  title        = {Pith review of: An Efficient and Globally Optimal Algorithm for Nonconvex QCQP with One Equality Constraint},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O3CVQFZN}},
  note         = {Machine review of arXiv:2506.03692}
}
read the original abstract

In this paper, we concentrate on a particular category of quadratically constrained quadratic programming (QCQP): nonconvex QCQP with one equality constraint. This type of QCQP problem optimizes a quadratic objective under a fixed second-order cost and has various engineering applications. It often serves as a subproblem in an iterative algorithm framework. However, the development of a high-quality and efficient solution remains an open problem in the existing literature. Traditionally, the Semidefinite Relaxation (SDR) technique is applied for an optimal solution with a prohibitively high order of time complexity. To improve computational efficiency, we propose a fast and non-iterative algorithm to reach a globally optimal solution. This algorithm consists of two consecutive stages: Simultaneous Diagonalization (SD) and Bisection Search (BS). The SD stage decouples the original problem through an affine mapping and the BS stage finds the optimal Lagrange multiplier by solving an equation induced from first- and second-order Karush-Kuhn-Tucker (KKT) conditions. In addition, we enrich the proposed algorithm with further extensions on the problem structure, namely, rank-deficient parameter, indefiniteness, constraint augmentation, and matrix-format variable. Numerical simulations show that the proposed algorithm achieves good numerical performance in terms of constraint satisfaction, optimality gap, and computational time, and scales to problem sizes at least ten times those supported by the traditional benchmarks.

Figures

Figures reproduced from arXiv: 2506.03692 by the authors.

Figure 1
Figure 1. An illustration plot of f (λ) = 1 (λ+5)2 + 4 (λ+2)2 + 1 (λ−1)2 + 4 (λ−4)2 + 1 (λ−7)2 . f (λ) indicates multiple KKT points in the first-order equation system. To figure out the optimal solution from the first-order KKT candidates, we need to introduce the second-order KKT condition [37]: ∂ 2L ∂y2 i ≥ 0, ∀i =⇒ 2 (ai + λ) ≥ 0, ∀i =⇒ λ ≥ − min ({ai}) (16) This is also a necessary condition for optimality. The intuition… view at source ↗
Figure 2
Figure 2. Performance metrics versus problem dimension [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Performance metrics versus problem dimension [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Performance metrics versus problem dimension [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Performance metrics versus problem dimension [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

45 extracted references · 39 canonical work pages

  1. [9]

    Generalizations of the trust region problem,

    J. J. Moré, “Generalizations of the trust region problem,” Optimization methods and Software, vol. 2, no. 3-4, pp. 189– 209, 1993

    work page 1993

  2. [30]

    A. R. Conn, N. I. Gould, and P. L. Toint,Trust region methods. SIAM, 2000

    2000

  3. [1]

    Eigenvalues and the max-cut prob- lem,

    B. Mohar and S. Poljak, “Eigenvalues and the max-cut prob- lem,”Czechoslovak Mathematical Journal, vol. 40, no. 2, pp. 343–352, 1990

    work page 1990

  4. [2]

    Quadratic knapsack problems,

    G. Gallo, P. L. Hammer, and B. Simeone, “Quadratic knapsack problems,” inCombinatorial optimization. Springer, 1980, pp. 132–149

    work page 1980

  5. [3]

    Trans- mit beamforming for physical-layer multicasting,

    N. D. Sidiropoulos, T. N. Davidson, and Z.-Q. Luo, “Trans- mit beamforming for physical-layer multicasting,”IEEE Trans. Signal Process., vol. 54, no. 6, pp. 2239–2251, 2006

    work page 2006

  6. [4]

    Far-field multicast beamforming for uniform linear antenna arrays,

    E. Karipidis, N. D. Sidiropoulos, and Z.-Q. Luo, “Far-field multicast beamforming for uniform linear antenna arrays,”IEEE Trans. Signal Process., vol. 55, no. 10, pp. 4916–4927, 2007

    work page 2007

  7. [5]

    Convex optimization-based beamforming,

    A. B. Gershman, N. D. Sidiropoulos, S. Shahbazpanahi, M. Bengtsson, and B. Ottersten, “Convex optimization-based beamforming,”IEEE Signal Process. Mag., vol. 27, no. 3, pp. 62–75, 2010

    work page 2010

  8. [6]

    Code design to optimize radar detection performance under ac- curacy and similarity constraints,

    A. De Maio, S. De Nicola, Y . Huang, S. Zhang, and A. Farina, “Code design to optimize radar detection performance under ac- curacy and similarity constraints,”IEEE Trans. Signal Process., vol. 56, no. 11, pp. 5618–5629, 2008

    work page 2008

Show all 45 references
  1. [7]

    I. T. Jolliffe,Principal Component Analysis. New York, NY , USA: Springer-Verlag, 2002

    2002

  2. [8]

    Relations between two sets of variates

    H. Hotelling, “Relations between two sets of variates.” Biometrika, vol. 28, no. 3-4, pp. 321–377, Dec. 1936

    1936

  3. [10]

    Majorization-minimization algorithms in signal processing, communications, and machine learning,

    Y . Sun, P. Babu, and D. P. Palomar, “Majorization-minimization algorithms in signal processing, communications, and machine learning,”IEEE Trans. Signal Process., vol. 65, no. 3, pp. 794– 816, 2016

    2016

  4. [11]

    Nonlinear programming,

    D. P. Bertsekas, “Nonlinear programming,”Journal of the Operational Research Society, vol. 48, no. 3, pp. 334–334, 1997

    1997

  5. [12]

    General heuristics for nonconvex quadrat- ically constrained quadratic programming,

    J. Park and S. Boyd, “General heuristics for nonconvex quadrat- ically constrained quadratic programming,”arXiv preprint arXiv:1703.07870, 2017

    2017 arXiv

  6. [13]

    S. Boyd, S. P. Boyd, and L. Vandenberghe,Convex optimization. Cambridge university press, 2004

    2004

  7. [14]

    Relaxations and randomized methods for nonconvex QCQPs,

    A. d’Aspremont and S. Boyd, “Relaxations and randomized methods for nonconvex QCQPs,”EE392 Class Notes, Stanford University, vol. 1, pp. 1–16, 2003

    2003

  8. [15]

    Quadratic programming is in NP,

    S. A. Vavasis, “Quadratic programming is in NP,”Information Processing Letters, vol. 36, no. 2, pp. 73–77, 1990

    1990

  9. [16]

    Feasible point pursuit and successive approximation of non-convex QCQPs,

    O. Mehanna, K. Huang, B. Gopalakrishnan, A. Konar, and N. D. Sidiropoulos, “Feasible point pursuit and successive approximation of non-convex QCQPs,”IEEE Signal Process. Lett., vol. 22, no. 7, pp. 804–808, 2014

    2014

  10. [17]

    First-order methods for fast feasibility pursuit of non-convex QCQPs,

    A. Konar and N. D. Sidiropoulos, “First-order methods for fast feasibility pursuit of non-convex QCQPs,”IEEE Trans. Signal Process., vol. 65, no. 22, pp. 5927–5941, 2017

    2017

  11. [18]

    A relax- ation method for nonconvex quadratically constrained quadratic programs,

    F. A. Al-Khayyal, C. Larsen, and T. Van V oorhis, “A relax- ation method for nonconvex quadratically constrained quadratic programs,”Journal of Global Optimization, vol. 6, no. 3, pp. 215–230, 1995

    1995

  12. [19]

    Exact solutions of some nonconvex quadratic optimization problems via SDP and SOCP relax- ations,

    S. Kim and M. Kojima, “Exact solutions of some nonconvex quadratic optimization problems via SDP and SOCP relax- ations,”Computational optimization and applications, vol. 26, pp. 143–154, 2003

    2003

  13. [20]

    Semidefinite relaxation of quadratic optimization problems,

    Z.-Q. Luo, W.-K. Ma, A. M.-C. So, Y . Ye, and S. Zhang, “Semidefinite relaxation of quadratic optimization problems,” IEEE Signal Process. Mag., vol. 27, no. 3, pp. 20–34, 2010

    2010

  14. [21]

    Interior-point algorithms for semidefinite pro- gramming problems derived from the KYP lemma,

    L. Vandenberghe, V . R. Balakrishnan, R. Wallin, A. Hansson, and T. Roh, “Interior-point algorithms for semidefinite pro- gramming problems derived from the KYP lemma,”Positive polynomials in control, pp. 195–238, 2005

    2005

  15. [22]

    Semidefinite programming,

    L. Vandenberghe and S. Boyd, “Semidefinite programming,” SIAM review, vol. 38, no. 1, pp. 49–95, 1996

    1996

  16. [23]

    Optimal estimation of sensor biases for asynchronous multi-sensor data fusion,

    W. Pu, Y .-F. Liu, J. Yan, H. Liu, and Z.-Q. Luo, “Optimal estimation of sensor biases for asynchronous multi-sensor data fusion,”Mathematical Programming, vol. 170, pp. 357–386, 2018

    2018

  17. [24]

    Rank-constrained separable semidefinite programming with applications to optimal beam- forming,

    Y . Huang and D. P. Palomar, “Rank-constrained separable semidefinite programming with applications to optimal beam- forming,”IEEE Trans. Signal Process., vol. 58, no. 2, pp. 664– 678, 2009

    2009

  18. [25]

    Consensus-ADMM for gen- eral quadratically constrained quadratic programming,

    K. Huang and N. D. Sidiropoulos, “Consensus-ADMM for gen- eral quadratically constrained quadratic programming,”IEEE Trans. Signal Process., vol. 64, no. 20, pp. 5297–5310, 2016

    2016

  19. [26]

    Fast approximation algo- rithms for a class of non-convex QCQP problems using first- order methods,

    A. Konar and N. D. Sidiropoulos, “Fast approximation algo- rithms for a class of non-convex QCQP problems using first- order methods,”IEEE Trans. Signal Process., vol. 65, no. 13, pp. 3494–3509, 2017

    2017

  20. [27]

    Gander,On the linear least squares problem with a quadratic constraint

    W. Gander,On the linear least squares problem with a quadratic constraint. Computer Science Department, Stanford University CA, 1978

    1978

  21. [28]

    Quadratically constrained least squares and quadratic problems,

    G. H. Golub and U. V on Matt, “Quadratically constrained least squares and quadratic problems,”Numerische Mathematik, vol. 59, no. 1, pp. 561–580, 1991

    1991

  22. [29]

    Hidden convexity in some nonconvex quadratically constrained quadratic programming,

    A. Ben-Tal and M. Teboulle, “Hidden convexity in some nonconvex quadratically constrained quadratic programming,” Mathematical Programming, vol. 72, no. 1, pp. 51–63, 1996

    1996

  23. [31]

    A review of trust region algorithms for optimiza- tion,

    Y .-x. Yuan, “A review of trust region algorithms for optimiza- tion,” inIciam, vol. 99, no. 1, 2000, pp. 271–282

    2000

  24. [32]

    Recent advances in trust region algorithms,

    ——, “Recent advances in trust region algorithms,”Mathemat- ical Programming, vol. 151, pp. 249–281, 2015

    2015

  25. [33]

    Duality and solutions for quadratic programming over single non- homogeneous quadratic constraint,

    J.-M. Feng, G.-X. Lin, R.-L. Sheu, and Y . Xia, “Duality and solutions for quadratic programming over single non- homogeneous quadratic constraint,”Journal of Global Opti- mization, vol. 54, pp. 275–293, 2012

    2012

  26. [34]

    Eigenvalue-based algorithm and analysis for nonconvex QCQP with one constraint,

    S. Adachi and Y . Nakatsukasa, “Eigenvalue-based algorithm and analysis for nonconvex QCQP with one constraint,”Mathemat- ical Programming, vol. 173, no. 1-2, pp. 79–116, 2019

    2019

  27. [35]

    On local non-global minimizers of quadratic optimization problem with a single quadratic constraint,

    A. Taati and M. Salahi, “On local non-global minimizers of quadratic optimization problem with a single quadratic constraint,”Numerical Functional Analysis and Optimization, vol. 41, no. 8, pp. 969–1005, 2020

    2020

  28. [36]

    On local minimizers of nonconvex homogeneous quadratically constrained quadratic optimization with at most two constraints,

    M. Song, H. Liu, J. Wang, and Y . Xia, “On local minimizers of nonconvex homogeneous quadratically constrained quadratic optimization with at most two constraints,”SIAM Journal on Optimization, vol. 33, no. 1, pp. 267–293, 2023

    2023

  29. [37]

    Nocedal and S

    J. Nocedal and S. J. Wright,Numerical optimization. Springer, 1999

    1999

  30. [38]

    R. B. Lehoucq, D. C. Sorensen, and C. Yang,ARPACK users’ guide: solution of large-scale eigenvalue problems with implic- itly restarted Arnoldi methods. SIAM, 1998

    1998

  31. [39]

    An algorithm for general- ized matrix eigenvalue problems,

    C. B. Moler and G. W. Stewart, “An algorithm for general- ized matrix eigenvalue problems,”SIAM Journal on Numerical Analysis, vol. 10, no. 2, pp. 241–256, 1973

    1973

  32. [40]

    Mean-reverting portfolio with budget constraint,

    Z. Zhao and D. P. Palomar, “Mean-reverting portfolio with budget constraint,”IEEE Trans. Signal Process., vol. 66, no. 9, pp. 2342–2357, 2018

    2018

  33. [41]

    Alternating minimization algorithms for hybrid precoding in millimeter wave MIMO systems,

    X. Yu, J.-C. Shen, J. Zhang, and K. B. Letaief, “Alternating minimization algorithms for hybrid precoding in millimeter wave MIMO systems,”IEEE J. Sel. Topics Signal Process., vol. 10, no. 3, pp. 485–500, 2016. 11

    2016

  34. [42]

    Joint target and transmitter localization with aoa and dtd using passive sensors,

    Z. Mao, H. Su, J. Song, and K. Han, “Joint target and transmitter localization with aoa and dtd using passive sensors,”IEEE Signal Process. Lett., 2024

    2024

  35. [43]

    Absil, R

    P.-A. Absil, R. Mahony, and R. Sepulchre,Optimization algo- rithms on matrix manifolds. Princeton University Press, 2009

    2009

  36. [44]

    R. A. Horn and C. R. Johnson,Matrix analysis. Cambridge university press, 2012

    2012

  37. [45]

    Diag (d1)0 0Diag (d 2) # ST A,3· ST A,2ST A,1 =TΛ 0TT A1 =S A,1

    D. S. Bernstein,Matrix mathematics: theory, facts, and formu- las. Princeton university press, 2009. APPENDIXA PROOF OFTHEOREM2 Proof:We undoA 1 withTT T and work on the constraint of QCQP (1) step by step: xH A1x=x T TTT x= TT x T TT x ,(49) 2bT 1 x= 2b T 1 T−T TT x= 2 T−1b1 ...

    2009

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.