scholarly journals Optimal Lower Bound for Generalized Median Problems in Metric Space

Author(s):  
Xiaoyi Jiang ◽  
Horst Bunke
2009 ◽  
Vol 80 (3) ◽  
pp. 486-497 ◽  
Author(s):  
ANTHONY WESTON

AbstractDetermining meaningful lower bounds on the supremal strict p-negative type of classes of finite metric spaces is a difficult nonlinear problem. In this paper we use an elementary approach to obtain the following result: given a finite metric space (X,d) there is a constant ζ>0, dependent only on n=∣X∣ and the scaled diameter 𝔇=(diamX)/min{d(x,y)∣x⁄=y} of X (which we may assume is >1), such that (X,d) has p-negative type for all p∈[0,ζ] and strict p-negative type for all p∈[0,ζ). In fact, we obtain A consideration of basic examples shows that our value of ζ is optimal provided that 𝔇≤2. In other words, for each 𝔇∈(1,2] and natural number n≥3, there exists an n-point metric space of scaled diameter 𝔇 whose supremal strict p-negative type is exactly ζ. The results of this paper hold more generally for all finite semi-metric spaces since the triangle inequality is not used in any of the proofs. Moreover, ζ is always optimal in the case of finite semi-metric spaces.


2016 ◽  
Vol 16 (3) ◽  
pp. 507-522 ◽  
Author(s):  
Yanhui Su ◽  
Lizhen Chen ◽  
Xianjuan Li ◽  
Chuanju Xu

AbstractThe Ladyženskaja–Babuška–Brezzi (LBB) condition is a necessary condition for the well-posedness of discrete saddle point problems stemming from discretizing the Stokes equations. In this paper, we prove the LBB condition and provide the (optimal) lower bound for this condition for the triangular spectral method proposed by L. Chen, J. Shen, and C. Xu in [3]. Then this lower bound is used to derive an error estimate for the pressure. Some numerical examples are provided to confirm the theoretical estimates.


1966 ◽  
Author(s):  
A. Charnes ◽  
M. L. Kirby ◽  
W. M. Raike

1994 ◽  
Vol 52 (6) ◽  
pp. 339 ◽  
Author(s):  
A. Bertoni ◽  
Carlo Mereghetti ◽  
Giovanni Pighizzini

2017 ◽  
Vol 2019 (22) ◽  
pp. 6924-6932 ◽  
Author(s):  
Christoph Aistleitner ◽  
Kamalakshya Mahatab ◽  
Marc Munsch

Abstract We prove that there are arbitrarily large values of t such that $|\zeta (1+it)| \geq e^{\gamma } (\log _{2} t +\log _{3} t) + \mathcal{O}(1)$. This essentially matches the prediction for the optimal lower bound in a conjecture of Granville and Soundararajan. Our proof uses a new variant of the “long resonator” method. While earlier implementations of this method crucially relied on a “sparsification” technique to control the mean-square of the resonator function, in the present paper we exploit certain self-similarity properties of a specially designed resonator function.


COMBINATORICA ◽  
1992 ◽  
Vol 12 (4) ◽  
pp. 389-410 ◽  
Author(s):  
Jin-Yi Cai ◽  
Martin F�rer ◽  
Neil Immerman

Sign in / Sign up

Export Citation Format

Share Document