-
New Analysis of Overlap** Schwarz Methods for Vector Field Problems in Three Dimensions with Generally Shaped Domains
Authors:
Duk-Soon Oh,
Shangyou Zhang
Abstract:
This paper introduces a novel approach to analyzing overlap** Schwarz methods for Nédélec and Raviart--Thomas vector field problems. The theory is based on new regular stable decompositions for vector fields that are robust to the topology of the domain. Enhanced estimates for the condition numbers of the preconditioned linear systems are derived, dependent linearly on the relative overlap betwe…
▽ More
This paper introduces a novel approach to analyzing overlap** Schwarz methods for Nédélec and Raviart--Thomas vector field problems. The theory is based on new regular stable decompositions for vector fields that are robust to the topology of the domain. Enhanced estimates for the condition numbers of the preconditioned linear systems are derived, dependent linearly on the relative overlap between the overlap** subdomains. Furthermore, we present the numerical experiments which support our theoretical results.
△ Less
Submitted 18 April, 2024;
originally announced April 2024.
-
Safety-Critical Control under Multiple State and Input Constraints and Application to Fixed-Wing UAV
Authors:
Donggeon David Oh,
Dongjae Lee,
H. ** Kim
Abstract:
This study presents a framework to guarantee safety for a class of second-order nonlinear systems under multiple state and input constraints. To facilitate real-world applications, a safety-critical controller must consider multiple constraints simultaneously, while being able to impose general forms of constraints designed for various tasks (e.g., obstacle avoidance). With this in mind, we first…
▽ More
This study presents a framework to guarantee safety for a class of second-order nonlinear systems under multiple state and input constraints. To facilitate real-world applications, a safety-critical controller must consider multiple constraints simultaneously, while being able to impose general forms of constraints designed for various tasks (e.g., obstacle avoidance). With this in mind, we first devise a zeroing control barrier function (ZCBF) using a newly proposed nominal evading maneuver. By designing the nominal evading maneuver to 1) be continuously differentiable, 2) satisfy input constraints, and 3) be capable of handling other state constraints, we deduce an ultimate invariant set, a subset of the safe set that can be rendered forward invariant with admissible control inputs. Thanks to the development of the ultimate invariant set, we then propose a safety-critical controller, which is a computationally tractable one-step model predictive controller (MPC) with guaranteed recursive feasibility. We validate the proposed framework in simulation, where a fixed-wing UAV tracks a circular trajectory while satisfying multiple safety constraints including collision avoidance, bounds on flight speed and flight path angle, and input constraints.
△ Less
Submitted 8 August, 2023;
originally announced August 2023.
-
Smoothers Based on Nonoverlap** Domain Decomposition Methods for $H(\mathbf{curl})$ Problems: A Numerical Study
Authors:
Duk-Soon Oh
Abstract:
This paper presents a numerical study on multigrid algorithms of $V$-cycle type for problems posed in the Hilbert space $H(\mathbf{curl})$ in three dimensions. The multigrid methods are designed for discrete problems originated from the discretization using the hexahedral Nédélec edge element of the lowest-order. Our suggested methods are associated with smoothers constructed by substructuring bas…
▽ More
This paper presents a numerical study on multigrid algorithms of $V$-cycle type for problems posed in the Hilbert space $H(\mathbf{curl})$ in three dimensions. The multigrid methods are designed for discrete problems originated from the discretization using the hexahedral Nédélec edge element of the lowest-order. Our suggested methods are associated with smoothers constructed by substructuring based on domain decomposition methods of nonoverlap** type. Numerical experiments to demonstrate the robustness and the effectiveness of the suggested algorithms are also provided.
△ Less
Submitted 4 September, 2022;
originally announced September 2022.
-
Multigrid methods for 3$D$ $H(\mathbf{curl})$ problems with nonoverlap** domain decomposition smoothers
Authors:
Duk-Soon Oh
Abstract:
We propose V--cycle multigrid methods for vector field problems arising from the lowest order hexahedral Nédélec finite element. Since the conventional scalar smoothing techniques do not work well for the problems, a new type of smoothing method is necessary. We introduce new smoothers based on substructuring with nonoverlap** domain decomposition methods. We provide the convergence analysis and…
▽ More
We propose V--cycle multigrid methods for vector field problems arising from the lowest order hexahedral Nédélec finite element. Since the conventional scalar smoothing techniques do not work well for the problems, a new type of smoothing method is necessary. We introduce new smoothers based on substructuring with nonoverlap** domain decomposition methods. We provide the convergence analysis and numerical experiments that support our theory.
△ Less
Submitted 11 May, 2022;
originally announced May 2022.
-
Elements in pointed invariant cones in Lie algebras and corresponding affine pairs
Authors:
Karl-Hermann Neeb,
Daniel Oeh
Abstract:
In this note we study in a finite dimensional Lie algebra ${\mathfrak g}$ the set of all those elements x for which the closed convex hull of the adjoint orbit contains no affine lines; this contains in particular elements whose adjoint orbits generates a pointed convex cone~$C_x$. Assuming that ${\mathfrak g}$ is admissible, i.e., contains a generating invariant convex subset not containing affin…
▽ More
In this note we study in a finite dimensional Lie algebra ${\mathfrak g}$ the set of all those elements x for which the closed convex hull of the adjoint orbit contains no affine lines; this contains in particular elements whose adjoint orbits generates a pointed convex cone~$C_x$. Assuming that ${\mathfrak g}$ is admissible, i.e., contains a generating invariant convex subset not containing affine lines, we obtain a natural characterization of such elements, also for non-reductive Lie algebras.
Motivated by the concept of standard (Borchers) pairs in QFT, we also study pairs $(x,h)$ of Lie algebra elements satisfying $[h,x]=x$ for which $C_x$ pointed. Given $x$, we show that such elements $h$ can be constructed in such a way that ${\rm ad} h$ defines a $5$-grading, and characterize the cases where we even get a $3$-grading.
△ Less
Submitted 14 October, 2021;
originally announced October 2021.
-
Controlled pattern formation of stochastic Cucker-Smale systems with network structures
Authors:
Young-Pil Choi,
Doeun Oh,
Oliver Tse
Abstract:
We present a new stochastic particle system on networks which describes the flocking behavior and pattern formation. More precisely, we consider Cucker-Smale particles with decentralized formation control and multiplicative noises on symmetric and connected networks. Under suitable assumptions on the initial configurations and the network structure, we establish time-asymptotic stochastic flocking…
▽ More
We present a new stochastic particle system on networks which describes the flocking behavior and pattern formation. More precisely, we consider Cucker-Smale particles with decentralized formation control and multiplicative noises on symmetric and connected networks. Under suitable assumptions on the initial configurations and the network structure, we establish time-asymptotic stochastic flocking behavior and pattern formation of solutions for the proposed stochastic particle system. Our approach is based on the Lyapunov functional energy estimates, and it does not require any spectral information of the graph associated with the network structure.
△ Less
Submitted 16 May, 2021;
originally announced May 2021.
-
Q-Rater: Non-Convex Optimization for Post-Training Uniform Quantization
Authors:
Byeongwook Kim,
Dongsoo Lee,
Yeonju Ro,
Yongkweon Jeon,
Se Jung Kwon,
Baeseong Park,
Daehwan Oh
Abstract:
Various post-training uniform quantization methods have usually been studied based on convex optimization. As a result, most previous ones rely on the quantization error minimization and/or quadratic approximations. Such approaches are computationally efficient and reasonable when a large number of quantization bits are employed. When the number of quantization bits is relatively low, however, non…
▽ More
Various post-training uniform quantization methods have usually been studied based on convex optimization. As a result, most previous ones rely on the quantization error minimization and/or quadratic approximations. Such approaches are computationally efficient and reasonable when a large number of quantization bits are employed. When the number of quantization bits is relatively low, however, non-convex optimization is unavoidable to improve model accuracy. In this paper, we propose a new post-training uniform quantization technique considering non-convexity. We empirically show that hyper-parameters for clip** and rounding of weights and activations can be explored by monitoring task loss. Then, an optimally searched set of hyper-parameters is frozen to proceed to the next layer such that an incremental non-convex optimization is enabled for post-training quantization. Throughout extensive experimental results using various models, our proposed technique presents higher model accuracy, especially for a low-bit quantization.
△ Less
Submitted 5 May, 2021;
originally announced May 2021.
-
Nets of standard subspaces induced by antiunitary representations of admissible Lie groups I
Authors:
Daniel Oeh
Abstract:
Let $(π, \mathcal{H})$ be a strongly continuous unitary representation of a 1-connected Lie group $G$ such that the Lie algebra $\mathfrak{g}$ of $G$ is generated by the positive cone $C_π:= \{x \in \mathfrak{g} : -i\partial π(x) \geq 0\}$ and an element $h$ for which the adjoint representation of $h$ induces a 3-grading of $\mathfrak{g}$. Moreover, suppose that $(π, \mathcal{H})$ extends to an an…
▽ More
Let $(π, \mathcal{H})$ be a strongly continuous unitary representation of a 1-connected Lie group $G$ such that the Lie algebra $\mathfrak{g}$ of $G$ is generated by the positive cone $C_π:= \{x \in \mathfrak{g} : -i\partial π(x) \geq 0\}$ and an element $h$ for which the adjoint representation of $h$ induces a 3-grading of $\mathfrak{g}$. Moreover, suppose that $(π, \mathcal{H})$ extends to an antiunitary representation of the extended Lie group $G_τ:= G \rtimes \{\mathbf{1}, τ_G\}$, where $τ_G$ is an involutive automorphism of $G$ with $\mathbf{L}(τ_G) = e^{iπ\mathrm{ad} h}$.
In a recent work by Neeb and Ólafsson, a method for constructing nets of standard subspaces of $\mathcal{H}$ indexed by open regions of $G$ has been introduced and applied in the case where $G$ is semisimple. In this paper, we extend this construction to general Lie groups $G$, provided the above assumptions are satisfied and the center of the ideal $\mathfrak{g}_C = C_π- C_π$ of $\mathfrak{g}$ is one-dimensional. The case where the center of $\mathfrak{g}_C$ has more than one dimension will be discussed in a separate paper.
△ Less
Submitted 3 September, 2021; v1 submitted 6 April, 2021;
originally announced April 2021.
-
Lie wedges of endomorphism semigroups of standard subspaces in admissible Lie algebras
Authors:
Daniel Oeh
Abstract:
Let $\mathfrak{g}$ be a real finite-dimensional Lie algebra containing pointed generating invariant closed convex cones. We determine those derivations $D$ of $\mathfrak{g}$ which induce a 3-grading of the form $\mathfrak{g} = \mathfrak{g}_{-1} \oplus \mathfrak{g}_0 \oplus \mathfrak{g}_1$ on $\mathfrak{g}$ such that the $(\pm 1)$-eigenspaces $\mathfrak{g}_{\pm 1}$ of $D$ are generated by the inter…
▽ More
Let $\mathfrak{g}$ be a real finite-dimensional Lie algebra containing pointed generating invariant closed convex cones. We determine those derivations $D$ of $\mathfrak{g}$ which induce a 3-grading of the form $\mathfrak{g} = \mathfrak{g}_{-1} \oplus \mathfrak{g}_0 \oplus \mathfrak{g}_1$ on $\mathfrak{g}$ such that the $(\pm 1)$-eigenspaces $\mathfrak{g}_{\pm 1}$ of $D$ are generated by the intersections with generating cones of the form $W = \mathcal{O}_f^*$, where $\mathcal{O}_f$ is the coadjoint orbit of a linear functional $f \in \mathfrak{z}(\mathfrak{g})^*$ and $\mathcal{O}_f^*$ is the dual cone of $\mathcal{O}_f$. In particular, we show that, if $\mathfrak{g}$ is solvable, no such derivation except the trivial one exists.
This continues our classification of Lie algebras generated by Lie wedges of endomorphism semigroups of standard subspaces. The classification is motivated by the relation of nets of standard subspaces to Haag-Kastler nets of von Neumann algebras in Algebraic Quantum Field Theory.
△ Less
Submitted 27 July, 2020;
originally announced July 2020.
-
Classification of 3-graded causal subalgebras of real simple Lie algebras
Authors:
Daniel Oeh
Abstract:
Let $(\mathfrak{g},τ)$ be a real simple symmetric Lie algebra and let $W \subset \mathfrak{g}$ be an invariant closed convex cone which is pointed and generating with $τ(W) = -W$. For elements $h \in \mathfrak{g}$ with $τ(h) = h$, we classify the Lie algebras $\mathfrak{g}(W,τ,h)$ which are generated by the closed convex cones \[C_{\pm}(W,τ,h) := (\pm W) \cap \mathfrak{g}_{\pm 1}^{-τ}(h),\] where…
▽ More
Let $(\mathfrak{g},τ)$ be a real simple symmetric Lie algebra and let $W \subset \mathfrak{g}$ be an invariant closed convex cone which is pointed and generating with $τ(W) = -W$. For elements $h \in \mathfrak{g}$ with $τ(h) = h$, we classify the Lie algebras $\mathfrak{g}(W,τ,h)$ which are generated by the closed convex cones \[C_{\pm}(W,τ,h) := (\pm W) \cap \mathfrak{g}_{\pm 1}^{-τ}(h),\] where $\mathfrak{g}^{-τ}_{\pm 1}(h) := \{x \in \mathfrak{g} : τ(x) = -x, [h,x] = \pm x\}$. These cones occur naturally as the skew-symmetric parts of the Lie wedges of endomorphism semigroups of certain standard subspaces. We prove in particular that, if $\mathfrak{g}(W,τ,h)$ is non-trivial, then it is either a hermitian simple Lie algebra of tube type or a direct sum of two Lie algebras of this type. Moreover, we give for each hermitian simple Lie algebra and each equivalence class of involutive automorphisms $τ$ of $\mathfrak{g}$ with $τ(W) = -W$ a list of possible subalgebras $\mathfrak{g}(W,τ,h)$ up to isomorphy.
△ Less
Submitted 24 October, 2020; v1 submitted 9 January, 2020;
originally announced January 2020.
-
Analytic extensions of representations of $*$-subsemigroups without polar decomposition
Authors:
Daniel Oeh
Abstract:
Let $(G,τ)$ be a finite-dimensional Lie group with an involutive automorphism $τ$ of $G$ and let $\mathfrak g = \mathfrak h \oplus \mathfrak q $ be its corresponding Lie algebra decomposition. We show that every non-degenerate strongly continuous representation on a complex Hilbert space $\mathcal H$ of an open $*$-subsemigroup $S \subset G$, where $s^* = τ(s)^{-1}$, has an analytic extension to a…
▽ More
Let $(G,τ)$ be a finite-dimensional Lie group with an involutive automorphism $τ$ of $G$ and let $\mathfrak g = \mathfrak h \oplus \mathfrak q $ be its corresponding Lie algebra decomposition. We show that every non-degenerate strongly continuous representation on a complex Hilbert space $\mathcal H$ of an open $*$-subsemigroup $S \subset G$, where $s^* = τ(s)^{-1}$, has an analytic extension to a strongly continuous unitary representation of the 1-connected Lie group $G_1^c$ with Lie algebra $[\mathfrak q,\mathfrak q] \oplus i\mathfrak q$.
We further examine the minimal conditions under which an analytic extension to the 1-connected Lie group $G^c$ with Lie algebra $\mathfrak h \oplus i\mathfrak q$ exists. This result generalizes the Lüscher-Mack Theorem and the extensions of the Lüscher-Mack Theorem for $*$-subsemigroups satisfying $S = S(G^τ)_0$ by Merigon, Neeb, and Ólafsson.
Finally, we prove that non-degenerate strongly continuous representations of certain $*$-subsemigroups $S$ can even be extended to representations of a generalized version of an Olshanski semigroup.
△ Less
Submitted 7 August, 2019; v1 submitted 27 December, 2018;
originally announced December 2018.
-
Laurent positivity of quantized canonical bases for quantum cluster varieties from surfaces
Authors:
So Young Cho,
Hyuna Kim,
Hyun Kyu Kim,
Doeun Oh
Abstract:
In 2006, Fock and Goncharov constructed a nice basis of the ring of regular functions on the moduli space of framed ${\rm PGL}_2$-local systems on a punctured surface $S$. The moduli space is birational to a cluster $\mathcal{X}$-variety, whose positive real points recover the enhanced Teichmüller space of $S$. Their basis is enumerated by integral laminations on $S$, which are collections of clos…
▽ More
In 2006, Fock and Goncharov constructed a nice basis of the ring of regular functions on the moduli space of framed ${\rm PGL}_2$-local systems on a punctured surface $S$. The moduli space is birational to a cluster $\mathcal{X}$-variety, whose positive real points recover the enhanced Teichmüller space of $S$. Their basis is enumerated by integral laminations on $S$, which are collections of closed curves in $S$ with integer weights. Around ten years later, a quantized version of this basis, still enumerated by integral laminations, was constructed by Allegretti and Kim. For each choice of an ideal triangulation of $S$, each quantum basis element is a Laurent polynomial in the exponential of quantum shear coordinates for edges of the triangulation, with coefficients being Laurent polynomials in $q$ with integer coefficients. We show that these coefficients are Laurent polynomials in $q$ with positive integer coefficients. Our result was expected in a positivity conjecture for framed protected spin characters in physics and provides a rigorous proof of it, and may also lead to other positivity results, as well as categorification. A key step in our proof is to solve a purely topological and combinatorial ordering problem about an ideal triangulation and a closed curve on $S$. For this problem we introduce a certain graph on $S$, which is interesting in its own right.
△ Less
Submitted 1 April, 2019; v1 submitted 17 October, 2017;
originally announced October 2017.
-
Multigrid Methods for Saddle Point Problems: Darcy Systems
Authors:
Susanne C. Brenner,
Duk-Soon Oh,
Li-yeng Sung
Abstract:
We design and analyze multigrid methods for the saddle point problems resulting from Raviart-Thomas-Nédélec mixed finite element methods (of order at least 1) for the Darcy system in porous media flow. Uniform convergence of the $W$-cycle algorithm in a nonstandard energy norm is established. Extensions to general second order elliptic problems are also addressed.
We design and analyze multigrid methods for the saddle point problems resulting from Raviart-Thomas-Nédélec mixed finite element methods (of order at least 1) for the Darcy system in porous media flow. Uniform convergence of the $W$-cycle algorithm in a nonstandard energy norm is established. Extensions to general second order elliptic problems are also addressed.
△ Less
Submitted 26 November, 2015;
originally announced November 2015.
-
MacWilliams-type equivalence relations
Authors:
Soohak Choi,
Jong Yoon Hyun,
Hyun Kwang Kim,
Dong Yeol Oh
Abstract:
Let $\mathcal{P}$ be a poset on $[n]$, $\mathcal{I}(\mathcal{P})$ the set of order ideals of $\mathcal{P}$ and $E$ an equivalence relation on $\mathcal{I}(\mathcal{P})$. The concepts of the dual relation $E^*$ of an equivalence relation $E$, the $E$-weight (resp. $E^*$-weight) distribution of a linear poset code (resp. its dual poset code) and a MacWilliams-type equivalence relation are introduced…
▽ More
Let $\mathcal{P}$ be a poset on $[n]$, $\mathcal{I}(\mathcal{P})$ the set of order ideals of $\mathcal{P}$ and $E$ an equivalence relation on $\mathcal{I}(\mathcal{P})$. The concepts of the dual relation $E^*$ of an equivalence relation $E$, the $E$-weight (resp. $E^*$-weight) distribution of a linear poset code (resp. its dual poset code) and a MacWilliams-type equivalence relation are introduced. We give a characterization for a MacWilliams-type equivalence relation in terms of MacWilliams-type identities for a linear poset code. Three kinds of equivalence relations on $\mathcal{I}(\mathcal{P})$ which are of MacWilliams-type are found, i.e., $(i)$ we show that every equivalence relation defined by the automorphism of $\mathcal{P}$ is a MacWilliams-type; $(ii)$ we provide a new characterization for poset structures when the equivalence relation defined by the same cardinality on $\mathcal{I}(\mathcal{P})$ becomes a MacWilliams-type; $(iii)$ we also give necessary and sufficient conditions for poset structures in which the equivalence relation defined by the order-isomorphism on $\mathcal{I}(\mathcal{P})$ is a MacWilliams-type.
△ Less
Submitted 31 December, 2012; v1 submitted 4 May, 2012;
originally announced May 2012.
-
Structures and lower bounds for binary covering arrays
Authors:
Soohak Choi,
Hyun Kwang Kim,
Dong Yeol Oh
Abstract:
A $q$-ary $t$-covering array is an $m \times n$ matrix with entries from $\{0, 1, ..., q-1\}$ with the property that for any $t$ column positions, all $q^t$ possible vectors of length $t$ occur at least once. One wishes to minimize $m$ for given $t$ and $n$, or maximize $n$ for given $t$ and $m$. For $t = 2$ and $q = 2$, it is completely solved by Rényi, Katona, and Kleitman and Spencer. They also…
▽ More
A $q$-ary $t$-covering array is an $m \times n$ matrix with entries from $\{0, 1, ..., q-1\}$ with the property that for any $t$ column positions, all $q^t$ possible vectors of length $t$ occur at least once. One wishes to minimize $m$ for given $t$ and $n$, or maximize $n$ for given $t$ and $m$. For $t = 2$ and $q = 2$, it is completely solved by Rényi, Katona, and Kleitman and Spencer. They also show that maximal binary 2-covering arrays are uniquely determined. Roux found the lower bound of $m$ for a general $t, n$, and $q$. In this article, we show that $m \times n$ binary 2-covering arrays under some constraints on $m$ and $n$ come from the maximal covering arrays. We also improve the lower bound of Roux for $t = 3$ and $q = 2$, and show that some binary 3 or 4-covering arrays are uniquely determined.
△ Less
Submitted 2 November, 2011;
originally announced November 2011.
-
Optimal codes in deletion and insertion metric
Authors:
Hyun Kwang Kim,
Joon Yop Lee,
Dong Yeol Oh
Abstract:
We improve the upper bound of Levenshtein for the cardinality of a code of length 4 capable of correcting single deletions over an alphabet of even size. We also illustrate that the new upper bound is sharp. Furthermore we will construct an optimal perfect code capable of correcting single deletions for the same parameters.
We improve the upper bound of Levenshtein for the cardinality of a code of length 4 capable of correcting single deletions over an alphabet of even size. We also illustrate that the new upper bound is sharp. Furthermore we will construct an optimal perfect code capable of correcting single deletions for the same parameters.
△ Less
Submitted 22 March, 2010; v1 submitted 20 October, 2008;
originally announced October 2008.
