Projects

December 21, 2014
0

中文论文

差异演化的实验研究. 控制与决策, 2004, 19(1): 49-52

摘要：首先基于一些实例研究了差异演化(DE)的参数选择问题;然后在分析DE特点的基础上,将缩放因子F由固定数值设为随机函数,实现了一个简化的DE版本(SDE).该方法不仅减少了需调整的参数,而且对CR的参数选择更为宽松.与已有文献中遗传算法的带约束型数值优化问题的实验结果对比,表明SDE能在较少的计算次数内获得较好的结果.
* 差异演化算法(Differential Evolution, DE)，又称为差分演化算法或差分进化算法 (相关论文和源代码)

微粒群算法综述. 控制与决策, 2003, 18(2): 129-134

摘要：讨论微粒群算法的开发与应用.首先回顾从1995年以来的开发过程,然后根据一些已有的测试结果对其参数设置进行系统地分析,并讨论一些非标准的改进手段,如簇分解、选择方法、邻域算子、无希望/重新希望方法等.介绍了一些常用的测试函数,以及与其他演化算法的比较.最后讨论了一些已经开发和在将来有希望的领域中的应用.
* 微粒群算法(Particle Swarm Optimization, PSO)，又称为粒子群优化算法、粒子群算法、或微粒群优化算法 (相关论文和源代码)

半导体器件建模与优化系统. 半导体学报, 2003, 24(3): 327-331

摘要：随着器件尺寸的缩小,器件特性空间变得越来越复杂.如果仍采用手工参数调整的方法,不仅需要有较好的器件物理知识,而且也不一定能得到合适的结果.为节约设计时间,对半导体器件建模与优化系统(MOSSED)进行了研究与实现.该系统可以对半导体器件进行有效地建模、优化和综合,以得到所需要的器件.通过一些实例对部分功能进行了说明,并和一些已有的系统进行了比较.

一种防止浮点遗传算法早熟收敛的父代选择策略. 控制与决策, 2002, 17(5): 625-628

摘要：通过对浮点遗传算法早熟收敛现象的分析,提出了一种新的父代选择策略,即使用当前代的子代个体作为下代的父代个体,可使交叉算子持续地探索和开发新空间.引入对个体的代数保护策略,即在它发生变异前保证有足够的演化,可以避免对新空间不成熟的开发.通过与其它父代选择策略的对比,并通过实验和GENOCOP系统比较,表明本方法能得到较好的结果.

半导体器件与工艺综合. 半导体学报, 2002, 23(7): 772-776

摘要：综合技术作为一项重要的研究方法，不仅在电路设计中获得广泛应用，而且在器件和工艺的研究上同样获得应用。利用器件与工艺综合的思想，我们开发出自顶向下的新的器件和工艺设计方法，和实现了该设计方法的MOSPAD 软件，并利用MOSPAD 系统做出一定的综合结果。本文分别做出关于器件与工艺综合的两个实例，即对FIB 器件的器件综合和对阱形成工艺模块进行的工艺综合，并证明了自顶向下的器件与工艺综合思想的可行性。

改进的用于半导体器件综合系统的遗传算法. 电子学报, 2002, 30(2): 184-186

摘要：本文对应用于器件综合系统的遗传算法GENOCOP进行了改进.将实数设计空间根据参数的工艺精度影响转换为整型空间,并加入适应性复合算子利用已经得到的点来扩展和开发准可行空间.使其保持有效搜索到可行解的特性的同时,在同等的算法设置下,提高了对可行空间的覆盖率(约2.87倍),可以帮助设计人员更有效的设计可工作的器件.

应用遗传算法实现MOS器件综合. 半导体学报, 2002, 23(1): 95-101

摘要：为实现器件综合，即从期望的器件性能出发得到优化的器件设计参数，最关键的是要选用有效的优化搜索算法。文章将遗传算法应用于实现一个器件综合的原型系统，并通过对FIBMOS器件的综合设计，验证了遗传算法和该器件综合原型系统的有效性。器件的参数化表示也被作为器件综合的重要问题进行了讨论。

October 17, 2014
0

Work with the Polymath project was published on Research in the Mathematical Sciences

Tags : polymath publication

Category : News , Projects

The work with the Polymath project was published on Research in the Mathematical Sciences.

D. H. J. Polymath, et al. Variants of the Selberg sieve, and bounded intervals containing many primes.Research in the Mathematical Sciences, 2014, 1(12): 1-83. [DOI]

Here is the list of Polymath8b authors (arranged in alphabetical order of surname): Ignace Bogaert, Aubrey de Grey, Gergely Harcos, Emmanuel Kowalski, Philippe Michel, James Maynard, Paul Nelson, Pace Nielsen, Eytan Paldi, Andrew V. Sutherland, Terence Tao, Xiao-Feng Xie

Abstract: For any $m \geq 1$ , let $H_m$ denote the quantity $\liminf_{n \to \infty} (p_{n+m}-p_n)$ , where $p_n$ is the $n^{\text{th}}$ prime. A celebrated recent result of Zhang showed the finiteness of $H_1$ , with the explicit bound $H_1 \leq 70000000$ . This was then improved by us (the Polymath8 project) to $H_1 \leq 4680$ , and then by Maynard to $H_1 \leq 600$ , who also established for the first time a finiteness result for $H_m$ for $m \geq 2$ , and specifically that $H_m \ll m^3 e^{4m}$ . If one also assumes the Elliott-Halberstam conjecture, Maynard obtained the bound $H_1 \leq 12$ , improving upon the previous bound $H_1 \leq 16$ of Goldston, Pintz, and Y{\i}ld{\i}r{\i}m, as well as the bound $H_m \ll m^3 e^{2m}$ . In this paper, we extend the methods of Maynard by generalizing the Selberg sieve further, and by performing more extensive numerical calculations. As a consequence, we can obtain the bound $H_1 \leq 246$ unconditionally, and $H_1 \leq 6$ under the assumption of the generalized Elliott-Halberstam conjecture. Indeed, under the latter conjecture we show the stronger statement that for any admissible triple $(h_1,h_2,h_3)$ , there are infinitely many $n$ for which at least two of $n+h_1,n+h_2,n+h_3$ are prime, and also obtain a related disjunction asserting that either the twin prime conjecture holds, or the even Goldbach conjecture is asymptotically true if one allows an additive error of at most $2$ , or both. We also modify the “parity problem” argument of Selberg to show that the $H_1 \leq 6$ bound is the best possible that one can obtain from purely sieve-theoretic considerations. For larger $m$ , we use the distributional results obtained previously by our project to obtain the unconditional asymptotic bound $H_m \ll m e^{(4-\frac{28}{157})m}$ , or $H_m \ll m e^{2m}$ under the assumption of the Elliott-Halberstam conjecture. We also obtain explicit upper bounds for $H_m$ when $m=2,3,4,5$ .

March 17, 2014
0

Work with the Polymath project was reported by Der Spiegel

Tags : media polymath

Category : News , Projects

This work with the Polymath8 project was to find narrow gaps between primes.

derspiegel …

Mal steuerte Xiao-Feng Xie einen Vorschlag bei, ein Robotikexperte aus Pittsburgh. Mal meldete sich Terence Tao aus Los Angeles, den einige für den brillantesten aller lebenden Mathematiker halten. Wieder andere Anregungen kamen von einem Anonymus, der sich nur als v08ltu zu erkennen gab.

…

Work on Smart Urban Traffic Control published on Transportation Research Part C

Category : Activities , Research

Schedule-driven intersection control (SchIC) is the core control engine (the brain) of the smart and scalable urban traffic control system.

Xiao-Feng Xie, S. Smith, Liang Lu, and G. Barlow. Schedule-driven intersection control. Transportation Research Part C, 24: 168-189, 2012.

@Article{Xie2012,
Title = {Schedule-driven intersection control},
Author = {Xiao-Feng Xie and S. Smith and Liang Lu and G. Barlow},
Journal = {Transportation Research Part C},
Year = {2012},
Pages = {168-189},
Volume = {24},
PDF={http://www.wiomax.com/team/xie/paper/TRC12.pdf},
Doi = {10.1016/j.trc.2012.03.004}
}

Model-based intersection optimization strategies have been widely investigated for distributed traffic signal control in road networks. Due to the form of ‘‘black-box’’ optimization that is typically assumed, a basic challenge faced by these strategies is the combinatorial nature of the problem that must be solved. The underlying state space is exponential in the number of time steps in the look-ahead optimization horizon at a given time resolution. In this paper, we present a schedule-driven intersection control strategy, called SchIC, which addresses this challenge by exploiting the structural information in non-uniformly distributed traffic flow. Central to our method is an alternative formulation of intersection control optimization as a scheduling problem, which effectively reduces the state space through use of an aggregate representation on traffic flow data in the prediction horizon. A forward recursive algorithm is proposed for solving the scheduling problem, which makes use of a dominance condition to efficiently eliminate most states at early stages. SchIC thus achieves near optimal solutions with a polynomial complexity in the prediction horizon, and is insensitive to the granularity of time resolution that is assumed. The performance of SchIC with respect to both intersection control and implicit coordination between intersections is evaluated empirically on two ideal scenarios and a real-world urban traffic network. Some characteristics and possible real-world extensions of SchIC are also discussed.

December 20, 2011
0

Solving the Magic Square Problem Using Modern Heuristic Methods

Tags : code optimization software

Category : News , Software

Ms_sf_2 This problem is to solve the Magic Square Problem (constrained and unconstrained versions), a combinatorial optimization problem, using modern heuristic methods.

Magic squares have been a source of fascination since ancient times, over 4,000 years. A magic square is a square matrix of size n, containing each of the numbers 1 to n² exactly once, in which each column, each row, and both diagonals add up to the same magic number.

It is possible to impose many different constraints on a standard magic square problem. Here the constrained version stipulates that the solution matrix must have a pre-defined contiguous sub-matrix.

Here is the binary code, and here is the readme file. As the 2nd runner-up, this program solved the constrained version of 400 x 400 magic square within a minute in the 2011 International Optimisation Competition.

X. Xie, “Meta-LS Solver for Magic Square Competition,” International Optimisation Competition, 2011.

@TechReport{xie2011tr00,
Title = {Meta-LS Solver for Magic Square Competition},
Author = {Xiao-Feng Xie},
Code={http://www.wiomax.com/team/xie/project/magic/MagicSquare.jar},
PDF={http://www.wiomax.com/team/xie/project/magic/IOC_readme.pdf},
Institution={International Optimisation Competition},
Year = {2011}
}

« Previous Page

Xiao-Feng Xie, Ph.D.

Category Archives: Projects

0

中文论文

0

Work with the Polymath project was published on Research in the Mathematical Sciences

0

Work with the Polymath project was reported by Der Spiegel

0

Work on Smart Urban Traffic Control published on Transportation Research Part C

0

Solving the Magic Square Problem Using Modern Heuristic Methods

Archives