FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1527166211> ?p ?o ?g. }

• W1527166211 abstract "Due to measurement uncertainty, after measuring a value of a physical quantity (or quantities), we do not get its exact value, we only get a set of possible values of this quantity (quantities). In case of 1-D quantities, we get an interval of possible values. It is known that the family of all real intervals is closed under point-wise arithmetic operations (+,−, ·) (i.e., this family forms an arithmetic). This closeness is efficiently used to estimate the set of possible values for y = f(x1, . . . , xn) from the known sets of possible values for xi. In some practical problems, physical quantities are complexvalued; it is therefore desirable to find a similar closed family (arithmetic) of complex sets. We follow K. Nickel’s 1980 paper to show that, in contrast to 1-D interval case, there is no finite-dimensional arithmetic. We prove this result by reformulating it as a geometric problem of finding a finite-dimensional family of planar sets which is closed under Minkowski addition, rotation, and dilation. Data processing: a practical problem which leads to arithmetic of complex sets. In many real-life situations, we are interested in the value of some physical quantity y which is difficult (or even impossible) to measure directly. To estimate y, we measure directly measurable quantities x1, . . . , xn which have a known relationship with y, and then reconstruct y from the results x1, . . . , xn of these measurements by using this known relation: ỹ = f(x1, . . . , xn), where f is a known algorithm. Measurements are never 100% accurate; as a result, the actual value xi of each measured quantity may differ from the measured value xi. If we know the upper bound ∆i for the measurement error |∆xi| = |xi − xi|, then after we get the measurement result xi, we can conclude that the actual value xi of the measured quantity belongs to the interval xi = [xi −∆i, xi +∆i]. A natural question is: when xi ∈ xi, what is the resulting interval y = f(x1, . . . ,xn) = {f(x1, . . . , xn) |xi ∈ xi} of possible values of y? Computing the exact bounds for the range interval is, in general, computationally difficult (see e.g., [Kreinovich et al. 1997]). However, there are efficient methods of computing an enclosure Y ⊇ y for this range; these methods are called methods of interval computations (see, e.g., [Hammer et al. 1993], [Hansen 1992], [Kearfott 1996], [Kearfott et al. 1996], [Moore 1979]). For example, we can use “naive interval computations”: describe the algorithm f as a sequence of elementary arithmetic operations (+,−, ·, /), and on each step, replace each operation ⊙ with numbers by the corresponding operation with intervals: x⊙ y = {x⊙ y |x ∈ x, y ∈ y}. (1) For intervals, we have explicit formulas for these arithmetic operations: e.g., [a, a] + [b, b] = [a+ b, a+ b], etc. For example, to estimate the range of the function f(x1) = x1 · (1 − x1), we describe the algorithm f as a sequence of two arithmetic operations: • computing the intermediate value r1 := 1− x1, and • computing the product f := x1 · r1. So, to estimate the range f([0, 1]), we compute r1 := 1 − [0, 1] = [0, 1], and then get the final enclosure Y := x1 · r1 = [0, 1] · [0, 1] = [0, 1] (this is, of course, a superset of the actual range [0,0.25]). Similar range estimation problems appear when the physical quantities are described by complex numbers. It is therefore desirable to find a similar technique for complex numbers. The methodology of naive interval computations is based on the fact that the set of all intervals (including degenerate intervals – real numbers) is closed under point-wise arithmetic operations (1) (except, of course, division by an interval y containing 0). In other words, arithmetic operations are well defined on the family of all intervals, so we can talk about the arithmetic of intervals. Hence, it is desirable to look for families of subsets of complex numbers which are also closed under arithmetic operations, i.e., to look for an arithmetic of complex sets. We want these subsets to be representable in a computer, where we can only store finitely many parameters and therefore, we want these sets to form a finite-dimensional (finite-parametric) family. Also, we want to take into consideration that real numbers are an important practical case of complex numbers; therefore, realline intervals (corresponding to imprecisely known real numbers) should be a particular case of this more general family of complex sets. Reasonable families of complex sets do not form a complex arithmetic: the empirical fact and the resulting question. There are several natural complex analogues of real-line intervals: • boxes, i.e., rectangular parallel to real axis; • ellipses (including real-line intervals as degenerate ellipses), etc. None of these families is closed under point-wise arithmetic operations (1). Moreover, they are not even closed under a limited set of arithmetic operations which includes addition and multiplication by complex numbers. A natural question is: Is there a finite-dimensional family of complex sets which is closed under these operations? To answer this question, let us reformulate it in geometric terms. Reformulating the question in geometric terms. In geometric terms, a complex plane is simply a plane, so we are looking for families of planar sets. The sum (1) of two planar sets is simply their Minkowski sum. In geometric terms, if we multiply a complex number t by another complex number z = ρ · exp(iφ), this mean that we first rotate t by an angle φ around the origin O = (0, 0) of the coordinate system, and then dilate the rotated point ρ times. Thus, the pointwise product z · T of a complex number z and a set T means that we first rotate the set T , and then dilate the result of this rotation. Hence, we arrive at the following definition: Definition. Let R be a plane. By an arithmetic of complex sets, we mean a family F of planar sets which satisfies the following three properties: • F contains all sub-intervals of the x-axis R× {0}; • F is closed under Minkowski addition, and • F is closed under rotations and dilations around O = (0, 0). A finite-dimensional family can be defined in a standard topological way: if we restrict ourselves to bounded and closed (hence, compact) sets, we can use Hausdorff distance between sets to define a topology; once the family is a topological space, we can use standard topological definitions to define its dimension. The question is: does there exist a finite-dimensional arithmetic of complex sets? Nickel’s answer, and why it is not final. In his paper [Nickel 1980], K. Nickel proves that “finite-dimensional” arithmetics of complex sets do not exist. However, in his formulation, he only considers sets with piece-wise smooth boundaries, and he uses a non-standard (and non-topological) definition of dimension. To be more precise, he calls a family “at least m-dimensional” if this family contains at least one set with m “corner” (nonsmooth) points, and he proves that every arithmetic of complex sets is “infinite-dimensional” in this sense by proving that it contains a m-cornered set Bm for each m. From the topological viewpoint, all these sets Bm form a family of dimension 0, and therefore, Nickel’s proof does not answer our question. Final answer. We will show that a minor modification of Nickel’s construction does lead to the final answer: Proposition. There exists no finite-dimensional arithmetic of" @default.
• W1527166211 created "2016-06-24" @default.
• W1527166211 creator A5061275811 @default.
• W1527166211 creator A5082379060 @default.
• W1527166211 date "1999-01-01" @default.
• W1527166211 modified "2023-09-27" @default.
• W1527166211 title "Arithmetic of Complex Sets: Nickel's Classical Paper Revisited from a Geometric Viewpoint" @default.
• W1527166211 cites W1572450310 @default.
• W1527166211 cites W1579653618 @default.
• W1527166211 cites W1580809199 @default.
• W1527166211 cites W2129480103 @default.
• W1527166211 cites W3146520007 @default.
• W1527166211 hasPublicationYear "1999" @default.
• W1527166211 type Work @default.
• W1527166211 sameAs 1527166211 @default.
• W1527166211 citedByCount "0" @default.
• W1527166211 crossrefType "journal-article" @default.
• W1527166211 hasAuthorship W1527166211A5061275811 @default.
• W1527166211 hasAuthorship W1527166211A5082379060 @default.
• W1527166211 hasConcept C114614502 @default.
• W1527166211 hasConcept C118615104 @default.
• W1527166211 hasConcept C136119220 @default.
• W1527166211 hasConcept C199343813 @default.
• W1527166211 hasConcept C202444582 @default.
• W1527166211 hasConcept C2777686260 @default.
• W1527166211 hasConcept C33923547 @default.
• W1527166211 hasConcept C71924100 @default.
• W1527166211 hasConcept C94375191 @default.
• W1527166211 hasConceptScore W1527166211C114614502 @default.
• W1527166211 hasConceptScore W1527166211C118615104 @default.
• W1527166211 hasConceptScore W1527166211C136119220 @default.
• W1527166211 hasConceptScore W1527166211C199343813 @default.
• W1527166211 hasConceptScore W1527166211C202444582 @default.
• W1527166211 hasConceptScore W1527166211C2777686260 @default.
• W1527166211 hasConceptScore W1527166211C33923547 @default.
• W1527166211 hasConceptScore W1527166211C71924100 @default.
• W1527166211 hasConceptScore W1527166211C94375191 @default.
• W1527166211 hasLocation W15271662111 @default.
• W1527166211 hasOpenAccess W1527166211 @default.
• W1527166211 hasPrimaryLocation W15271662111 @default.
• W1527166211 hasRelatedWork W1616376819 @default.
• W1527166211 hasRelatedWork W1987792882 @default.
• W1527166211 hasRelatedWork W2064868327 @default.
• W1527166211 hasRelatedWork W2069655509 @default.
• W1527166211 hasRelatedWork W2073800363 @default.
• W1527166211 hasRelatedWork W2087071857 @default.
• W1527166211 hasRelatedWork W2123005834 @default.
• W1527166211 hasRelatedWork W2135070950 @default.
• W1527166211 hasRelatedWork W220189161 @default.
• W1527166211 hasRelatedWork W2407936537 @default.
• W1527166211 hasRelatedWork W2611992761 @default.
• W1527166211 hasRelatedWork W2804020193 @default.
• W1527166211 hasRelatedWork W2949741879 @default.
• W1527166211 hasRelatedWork W2953117950 @default.
• W1527166211 hasRelatedWork W2963356545 @default.
• W1527166211 hasRelatedWork W2964204138 @default.
• W1527166211 hasRelatedWork W3089561538 @default.
• W1527166211 hasRelatedWork W3152090362 @default.
• W1527166211 hasRelatedWork W32133377 @default.
• W1527166211 hasRelatedWork W1181004388 @default.
• W1527166211 isParatext "false" @default.
• W1527166211 isRetracted "false" @default.
• W1527166211 magId "1527166211" @default.
• W1527166211 workType "article" @default.