中国科学院数学与系统科学研究院期刊网

15 May 1979, Volume 22 Issue 3
    

  • Select all
    |
    论文
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 261-275. https://doi.org/10.12386/A1979sxxb0028
    Abstract ( )   Knowledge map   Save
    The minimax optimization of unimodal functions of one variable with time delay has practical value in the allocation of experiments that need a longer time interval to turn out their results. The optimal strategies suggested thus far are only for the periods not more than three. The upper bound A_n (τ) of the length of feasible interval was presented in [2] for any period of time delay. It is proved in this paper that, even when the period is four, the upper bound A_n (4) is irreachable. A pseudooptimal strategy approximate to the upper bound is suggestd, and its error is estimated By using the approach of this paper, it can be proved that A_n (5) is also irreachable.
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 276-290. https://doi.org/10.12386/A1979sxxb0029
    Abstract ( )   Knowledge map   Save
    Let R be a denmnerable set, we say P = (p_(ij), i, j ∈R) is a stochastic matrix, if If h is a nonnegative function defined on R and Ph = h, then we say, h is a harmonic function. In this paper, the construction of harmonic functions of some stochastic matrixes (for example, branching stochastic matrixes and renewal stochastic matrixes) is investigated.
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 291-302. https://doi.org/10.12386/A1979sxxb0030
    Abstract ( )   Knowledge map   Save
    It is known that a Coxeter element w of the Weyl group W of a simple Lie algebrag has the following elegant property: If w is a Coxeter element of W, then itseigenvalues are just e~mj(2πi/h) (j = 1,2,…,l), where m_j(j= 1, 2,…, l) are the Poin care exponents of g,and h = 1 + o, where o is the order of the highest root of g.This note aims at proving the followingTheorem: w ∈W is a Coxeter element of W if and only if it satisfies: (1) 1 is not the eigenvalue of w, and, (2) the number of the cycles of w is at most l.Using similar method, we also give a simple proof of the two important results due to B. Kostant.
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 303-315. https://doi.org/10.12386/A1979sxxb0031
    Abstract ( )   Knowledge map   Save
    A ring Ω is called a complete ring of linear transformations if there exists a vector space A over a division ring such that Ω is the ring of all linear transformations of A Let be the socle of Ω. Then a subset {E_i} r of is called a correlative basis of if and only if either where, E_i~2=E_i,E_iE_j = 0, i≠j, i, j ∈F,Definition 1. A ring Ω is called a normal ring if and only if Ω is the complete ring of all linear transformations and for every correlative basis {E_i} r of.Definition 2.A primitive ring R is called normalizable if and only if there exists a normal ring Ω such that R Ω and the socle of Ω contains that of R.Definition 3. A chain of rings Ω_o Ω_1 …Ω_a …is called a standard ascending chain if it satisfies the following conditions:(i) for every ordinal number a there corresponds a component Ω_a of this chain.(ii) every component Ω_a of this chain is the complete ring of all linear transformations such that Ω_a is dense in Ω_(a+1).(iii) let be the socle of Ω_a, then …(iv) if a<β then Ω_βL_α is a minimal left ideal of Ω_β for every minimal left ideal L_a of Ω_a.Definition 4.A standard ascending chain of rings Ω_o Ω_1 …Ω_a… is called a non-set chain if Ω_a is a non-set, where ∪ is the set sum.Definition 5.A primitive ring R is called a ring having a non-set standard chain of rings Ω_o Ω_1 …Ω_a …if this chain is a non-set standard chain of rings and R is dense in Ω_o.Theorem Ⅰ. Any primitive ring is either a normalizable ring or a ring having a non-set standard chain of rings.Definition 6. A matrix ring M over a division ring K is called a matrix ring having a finite number of columns with non-zero eatries if and only if for every element r of M there corresponds n(r) number of cohunns of r such that all entries outside these n(r) columns are zeros, where n(r) is a positive integer.Theorem Ⅱ; Let R be a normalizable primitive ring, and I be any subset of elements of the soele of R with cardinal number Then the subring generated by I is isomorphic to a subring of matrix ring having a finite number of columns with non-zero entries over a division ring.
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 316-343. https://doi.org/10.12386/A1979sxxb0032
    Abstract ( )   Knowledge map   Save
    Let M~n be a compact C~∞ Riemann manifold of dimension n 2. Consider an arbitrarily given C~1 differential system S on M~n. Then S induces a C~1 one-parame-ter transformation group φ_t:M~n→M~n (-∞0 and T>0 such that It is also easily seen that, when S ∈The result which we obtain is the following theorem. It is the basic property of, as mentioned in the title.Theorem. There exists an open covering of, and corresponding to each H∈there exist numbcrs η_H>0 and T_H>0 sueh that: if V∈and S∈V, then Whenever a point on a peridic orbit of S and T_v≤t<∞, we haveWhenever is a periodic orbit of S with period To, x∈P,and 0=to
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 344-353. https://doi.org/10.12386/A1979sxxb0033
    Abstract ( )   Knowledge map   Save
    In 1974, N. Levinson proved his famous theorem: More than one third of zeros of Riemann's Zeta-function are on σ = 1/2. The key to realizing his method of proving this theorem was to calculate the following integralWhere U=TL~(-10),L=log T/2π,0
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 354-361. https://doi.org/10.12386/A1979sxxb0034
    Abstract ( )   Knowledge map   Save
    At present, we know that minimization methods realizing an arbitrarily given switching function by an and-or network are mainly the tabular method and the consensus method [1]. It is the aim of this paper to establish a new minimization method by combining the tabular method with the consensus method.
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 362-374. https://doi.org/10.12386/A1979sxxb0035
    Abstract ( )   Knowledge map   Save
    Let be the p-adic generalized Walsh func tions, where are the expresSions of p-adic numbers of bers of t,x,respectively The Walsh transform f of ∈L_(o,∞_~2 is defined by f(t)= and the Walsh transform on L_(o.∞) is defined similarly. It is known that for Walsh transform many basic properties analogous to Fourier transform are valid. In this paper we shall consider other properties. Introducing the function class of W_N -type and the concept of generalized logical derivatives, we Deal with the descent order of finite Walsh transform and obtain certain asymptotic behavoir of L(t). We Also give the necessary and sufficient conditions of functions which may be expressed by the form of (1). Finally, we study what role the function φ(x) and its derivative will play in the solution of an extremum problem, and the condition characterizing the best solution is obtained. Thus a model on a logical differencial-integral equation in connection with an interesting variation problem is immediately in sight.
  • Acta Mathematica Sinica, Chinese Series. 1979, 22(3): 375-388. https://doi.org/10.12386/A1979sxxb0036
    Abstract ( )   Knowledge map   Save
    Let f(x) ∈C[a, b]. It is well known that Schoenberg-Marsden's variation-diminishing spline approximation is given by the followingLet S~k(f) (x) = S(S~(k-1)(f))(x), k > 1. We have the iterated esquence {S~k(f) (x)}. In this paper, we prove that S~k(f)(x) converge uniformly to F(x) as k→∞. Here, F(x) is a polygonal function connected by the points of (x_i),f(x_1), i = 1,…, r, for all knots of x's which are of multiplicity m-1 or m.Obviously, The iterated resalts of V. D. approximations such as those of Bernstein (Bezier) polynomial and simple (continuous) splines are all the special case of our general considerration.