arXiv subjects

Makoto Yamashita

Publications and source records attributed to Makoto Yamashita.

At least 55 records · Page 3Linked to original sources

arXiv2018

A dual spectral projected gradient method for log-determinant semidefinite problems

We extend the result on the spectral projected gradient method by Birgin et al. in 2000 to a log-determinant semidefinite problem (SDP) with linear constraints and propose a spectral projected gradient method for the dual problem. Our method is based on alternate projections on the intersection of two convex sets, which first projects onto the box constraints and then onto a set defined by a linear matrix inequality. By exploiting structures of the two projections, we show the same convergence properties can be obtained for the proposed method as Birgin's method where the exact orthogonal projection onto the intersection of two convex sets is performed. Using the convergence properties, we prove that the proposed algorithm attains the optimal value or terminates in a finite number of iterations. The efficiency of the proposed method is illustrated with the numerical results on randomly generated synthetic/deterministic data and gene expression data, in comparison with other methods including the inexact primal-dual path-following interior-point method, the adaptive spectral projected gradient method, and the adaptive Nesterov's smooth method. For the gene expression data, our results are compared with the quadratic approximation for sparse inverse covariance estimation method. We show that our method outperforms the other methods in obtaining a better optimal value fast.

math.OC↗

arXiv2018

Polyhedral-based Methods for Mixed-Integer SOCP in Tree Breeding

Optimal contribution selection (OCS) is a mathematical optimization problem that aims to maximize the total benefit from selecting a group of individuals under a constraint on genetic diversity. We are specifically focused on OCS as applied to forest tree breeding, when selected individuals will contribute equally to the gene pool. Since the diversity constraint in OCS can be described with a second-order cone, equal deployment in OCS can be mathematically modeled as mixed-integer second-order cone programming (MI-SOCP). If we apply a general solver for MI-SOCP, non-linearity embedded in OCS requires a heavy computation cost. To address this problem, we propose an implementation of lifted polyhedral programming (LPP) relaxation and a cone-decomposition method (CDM) to generate effective linear approximations for OCS. In particular, CDM successively solves OCS problems much faster than generic approaches for MI-SOCP. The approach of CDM is not limited to OCS, so that we can also apply the approach to other MI-SOCP problems.

math.OC↗

arXiv2018

Tube representations and twisting of graded categories

We study deformation of tube algebra under twisting of graded monoidal categories. When a tensor category $\mathcal{C}$ is graded over a group $Γ$, a torus-valued 3-cocycle on $Γ$ can be used to deform the associator of $\mathcal{C}$. Based on a natural Fell bundle structure of the tube algebra over the action groupoid of the adjoint action of $Γ$, we show that the tube algebra of the twisted category is a 2-cocycle twisting of the original one.

math.QA↗

arXiv2018

Solving Pooling Problems by LP and SOCP Relaxations and Rescheduling Methods

The pooling problem is an important industrial problem in the class of network flow problems for allocating gas flow in pipeline transportation networks. For P-formulation of the pooling problem with time discretization, we propose second order cone programming (SOCP) and linear programming (LP) relaxations and prove that they obtain the same optimal value as the semidefinite programming relaxation. The equivalence among the optimal values of the three relaxations is also computationally shown. Moreover, a rescheduling method is proposed to efficiently refine the solution obtained by the SOCP or LP relaxation. The efficiency of the SOCP and the LP relaxation and the proposed rescheduling method is illustrated with numerical results on the test instances from the work of Nishi in 2010, some large instances, and Foulds 3, 4, 5 test problems.

math.OC↗

arXiv2018

A few remarks on the tube algebra of a monoidal category

We prove two results on the tube algebras of rigid C$^*$-tensor categories. The first is that the tube algebra of the representation category of a compact quantum group $G$ is a full corner of the Drinfeld double of $G$. As an application we obtain some information on the structure of the tube algebras of the Temperley-Lieb categories $TL(d)$ for $d>2$. The second result is that the tube algebras of weakly Morita equivalent C$^*$-tensor categories are strongly Morita equivalent. The corresponding linking algebra is described as the tube algebra of the $2$-category defining the Morita context.

math.OA↗

arXiv2018

Graded twisting of comodule algebras and module categories

Continuing our previous work on graded twisting of Hopf algebras and monoidal categories, we introduce a graded twisting construction for equivariant comodule algebras and module categories. As an example we study actions of quantum subgroups of $G\subset SL_{-1}(2)$ on $K_{-1}[x,y]$ and show that in most cases the corresponding invariant rings $K_{-1}[x,y]^G$ are invariant rings $K[x,y]^{G'}$ for the action of a classical subgroup $G'\subset SL(2)$. As another example we study Poisson boundaries of graded twisted categories and show that under the assumption of weak amenability they are graded twistings of the Poisson boundaries.

math.QA↗

arXiv2018

Ribbon braided module categories, quantum symmetric pairs and Knizhnik-Zamolodchikov equations

Let $\mathfrak u$ be a compact semisimple Lie algebra, and $σ$ be a Lie algebra involution of $\mathfrak u$. Let Rep$_q(\mathfrak u)$ be the ribbon braided tensor C*-category of $U_q(\mathfrak u)$-representations for $0<q<1$. We introduce three module C*-categories over Rep$_q(\mathfrak u)$ starting from the input data $(\mathfrak u,σ)$. The first construction is based on the theory of cyclotomic KZ-equations. The second construction uses the notion of quantum symmetric pair as developed by G. Letzter. The third construction uses a variation of Drinfeld twisting. In all three cases the module C*-category is ribbon twist-braided in the sense of A. Brochier---this is essentially due to B. Enriquez in the first case, is proved by S. Kolb in the second case, and is closely related to work of J. Donin, P. Kulish, and A. Mudrov in the third case. We formulate a conjecture concerning equivalence of these ribbon twist-braided module C*-categories, and confirm it in the rank one case.

math.QA↗

arXiv2017

Tracing cyclic homology pairings under twisting of graded algebras

We give a description of cyclic cohomology and its pairing with K-groups for 2-cocycle deformation of algebras graded over discrete groups. The proof relies on a realization of monodromy for the Gauss-Manin connection on periodic cyclic cohomology in terms of the cup product action of group cohomology.

math.QA↗

arXiv2017

Poisson boundaries of monoidal categories

Given a rigid C*-tensor category C with simple unit and a probability measure $μ$ on the set of isomorphism classes of its simple objects, we define the Poisson boundary of $(C,μ)$. This is a new C*-tensor category P, generally with nonsimple unit, together with a unitary tensor functor $Π: C \to P$. Our main result is that if P has simple unit (which is a condition on some classical random walk), then $Π$ is a universal unitary tensor functor defining the amenable dimension function on C. Corollaries of this theorem unify various results in the literature on amenability of C*-tensor categories, quantum groups, and subfactors.

math.OA↗

arXiv2017

Conic relaxation approaches for equal deployment problems

An important problem in the breeding of livestock, crops, and forest trees is the optimum of selection of genotypes that maximizes genetic gain. The key constraint in the optimal selection is a convex quadratic constraint that ensures genetic diversity, therefore, the optimal selection can be cast as a second-order cone programming (SOCP) problem. Yamashita et al. (2015) exploits the structural sparsity of the quadratic constraints and reduces the computation time drastically while attaining the same optimal solution. This paper is concerned with the special case of equal deployment (ED), in which we solve the optimal selection problem with the constraint that contribution of genotypes must either be a fixed size or zero. This involves a nature of combinatorial optimization, and the ED problem can be described as a mixed-integer SOCP problem. In this paper, we discuss conic relaxation approaches for the ED problem based on LP (linear programming), SOCP, and SDP (semidefinite programming). We analyze theoretical bounds derivedfrom the SDP relaxation approaches using the work of Tseng (2003) and show that the theoretical bounds are not quite sharp for tree breeding problems. We propose a steepest-ascent method that combines the solution obtained from the conic relaxation problems with a concept from discrete convex optimization in order to acquire an approximate solution for the ED problem in a practical time. From numerical tests, we observed that among the LP, SOCP, and SDP relaxation problems, SOCP gave a suitable solution from the viewpoints of the optimal values and the computation time. The steepest-ascent method starting from the SOCP solution provides high-quality solutions much faster than an existing method that has been widely used for the optimal selection problems and a branch-and-bound method.

math.OC↗

arXiv2017

Monodromy of the Gauss-Manin connection for deformation by group cocycles

We consider the 2-cocycle deformation of algebras graded by discrete groups. The action of the Maurer-Cartan form on cyclic cohomology is shown to be cohomologous to the cup product action of the group cocycle. This allows us to compute the monodromy of the Gauss-Manin connection in the strict deformation setting.

math.KT↗

arXiv2016

A Successive LP Approach with C-VaR Type Constraints for IMRT Optimization

Radiation therapy is considered to be one of important treatment protocols for cancers. Radiation therapy employs several beams of ionizing radiation to kill cancer tumors, but such irradiation also causes damage to normal tissues. Therefore, a treatment plan should satisfy dose-volume constraints (DVCs). Intensity-modulated radiotherapy treatment (IMRT) enables to control the beam intensities and gives more flexibility for the treatment plan to satisfy the DVCs. Romeijn et al. (2003) replaced the DVCs in an IMRT optimization with C-VaR (Conditional Value-at-Risk) type constraints, and proposed a numerical method based on linear programming (LP). Their approach reduced the computation cost of the original DVCs, but the feasible region of their LP problems was much narrow compared to the DVCs, therefore, their approach often failed to find a feasible plan even when the DVCs were not so tight. In this paper, we propose a successive LP approach with the C-VaR type constraints. We detect outliers form the solution of LP problems, and remove them from the domain of the C-VaR type constraints. This eases the sensitivity of C-VaR type constraints to outliers and we can search feasible plans from wider regions. Furthermore, we can give a mathematical proof that if the optimal value of the LP problem in the proposed approach is non-positive, the corresponding optimal solution satisfies all the DVCs. From numerical experiments on test data sets, we observed that our approach found feasible solutions more appropriately than existing LP approaches. In addition, our approach required fewer LP problems, and this led to a short computation time.

math.OC↗

arXiv2016

Drinfeld center and representation theory for monoidal categories

Motivated by the relation between the Drinfeld double and central property (T) for quantum groups, given a rigid C*-tensor category C and a unitary half-braiding on an ind-object, we construct a *-representation of the fusion algebra of C. This allows us to present an alternative approach to recent results of Popa and Vaes, who defined C*-algebras of monoidal categories and introduced property (T) for them. As an example we analyze categories C of Hilbert bimodules over a II$_1$-factor. We show that in this case the Drinfeld center is monoidally equivalent to a category of Hilbert bimodules over another II$_1$-factor obtained by the Longo-Rehren construction. As an application, we obtain an alternative proof of the result of Popa and Vaes stating that property (T) for the category defined by an extremal finite index subfactor $N \subset M$ is equivalent to Popa's property (T) for the corresponding SE-inclusion of II$_1$-factors. In the last part of the paper we study Müger's notion of weakly monoidally Morita equivalent categories and analyze the behavior of our constructions under the equivalence of the corresponding Drinfeld centers established by Schauenburg. In particular, we prove that property (T) is invariant under weak monoidal Morita equivalence.

math.OA↗

arXiv2016

Graded twisting of categories and quantum groups by group actions

Given a Hopf algebra $A$ graded by a discrete group together with an action of the same group preserving the grading, we define a new Hopf algebra, which we call the graded twisting of $A$. If the action is adjoint, this new Hopf algebra is a twist of $A$ by a pseudo-$2$-cocycle. Analogous construction can be carried out for monoidal categories. As examples we consider graded twistings of the Hopf algebras of nondegenerate bilinear forms, their free products, hyperoctahedral quantum groups and $q$-deformations of compact semisimple Lie groups. As applications, we show that the analogues of the Kazhdan-Wenzl categories in the general semisimple case cannot be always realized as representation categories of compact quantum groups, and for genuine compact groups, we analyze quantum subgroups of the new twisted compact quantum groups, providing a full description when the twisting group is cyclic of prime order.

math.QA↗

arXiv2016

Spin-1 Quantum Walks

We study the quantum walks of two interacting spin-1 bosons. We derive an exact solution for the time-dependent wave function that describes the two-particle dynamics governed by the one-dimensional spin-1 Bose-Hubbard model. We show that propagation dynamics in real space and mixing dynamics in spin space are correlated via the spin-dependent interaction in this system. The spin-mixing dynamics has two characteristic frequencies in the limit of large spin-dependent interactions. One of the characteristic frequencies is determined by the energy difference between two bound states, and the other frequency relates to the cotunneling process of a pair of spin-1 bosons. Furthermore, we numerically analyze the growth of the spin correlations in quantum walks. We find that long-range spin correlations emerge showing a clear dependence on the sign of the spin-dependent interaction and the initial state.

quant-ph↗

arXiv2016

Towards a classification of compact quantum groups of Lie type

This is a survey of recent results on classification of compact quantum groups of Lie type, by which we mean quantum groups with the same fusion rules and dimensions of representations as for a compact connected Lie group $G$. The classification is based on a categorical duality for quantum group actions recently developed by De Commer and the authors in the spirit of Woronowicz's Tannaka--Krein duality theorem. The duality establishes a correspondence between the actions of a compact quantum group $H$ on unital C$^*$-algebras and the module categories over its representation category Rep $H$. This is further refined to a correspondence between the braided-commutative Yetter--Drinfeld $H$-algebras and the tensor functors from Rep $H$. Combined with the more analytical theory of Poisson boundaries, this leads to a classification of dimension-preserving fiber functors on the representation category of any coamenable compact quantum group in terms of its maximal Kac quantum subgroup, which is the maximal torus for the $q$-deformation of $G$ if $q\ne1$. Together with earlier results on autoequivalences of the categories Rep $G_q$, this allows us to classify up to isomorphism a large class of quantum groups of $G$-type for compact connected simple Lie groups $G$. In the case of $G=SU(n)$ this class exhausts all non-Kac quantum groups.

math.QA↗

arXiv2016

Dynamical instability in the S=1 Bose-Hubbard model

We study the dynamical instabilities of superfluid flows in the S=1 Bose-Hubbard model. The time evolution of each spin component in a condensate is calculated based on the dynamical Gutzwiller approximation for a wide range of interactions, from a weakly correlated regime to a strongly correlated regime near the Mott-insulator transition. Owing to the spin-dependent interactions, the superfluid flow of the spin-1 condensate decays at a different critical momentum from a spinless case when the interaction strength is the same. We furthermore calculate the dynamical phase diagram of this model and clarify that the obtained phase boundary has very different features depending on whether the average number of particles per site is even or odd. Finally, we analyze the density and spin modulations that appear in association with the dynamical instability. We find that spin modulations are highly sensitive to the presence of a uniform magnetic field.

cond-mat.quant-gas↗

arXiv2015

BCS superconducting transitions in lattice fermions

We develop a general description of the superconductivity of lattice fermions based on the BCS theory. We propose a modeling of the density of states (DOS) of lattice fermions, where divergent and semi-metallic structures are described by asymptotic expansions around the Fermi energy. This modeling leads to a unified representation of the transition temperature $T_c$ at half filling, which reproduces asymptotic forms of $T_c$ derived in several lattices, such as the square, honeycomb, and Lieb lattices, for the weakly interacting limit. The derived asymptotic forms of $T_c$ are categorized into four types, which is attributed to the different responses of the degenerate fermions depending on the DOS structures. The DOS with a delta-functional singularity induces the highest $T_c$ in the weakly interacting region, where $T_c$ is linearly proportional to the pairing interaction $U$. Three kinds of universal ratios defined in the BCS theory no longer reduce to constants independent of the system parameters but can be parameterized with a certain variable that characterizes the singular structures of the DOS. We find universal relationship among thermodynamic quantities that holds for all parameter regions. Further, we numerically demonstrate that in multi-band systems the correlation effects can induce an effective delta-functional singularity. This phenomenon generally appears in multi-energy (or multi-gap) systems and may provide a plausible guideline for material designs of high-$T_c$ superconductor.

cond-mat.supr-con↗