from    
to    
search  

 


硬X射线谱学技术发展和应用
可再生能源科学沙龙 第一次校友论坛:高可再生能源比例下的新型电力系统
全球变化科学紫荆论坛第426期:基于气候相似性分析的台风特征及致灾研究
C3PU: An Automated Portable Platform for Organic Synthesis
报告题目:
Optimal Bounds for Predecessor Search and the First Separation between Linear and Polynomial Space
 报告人:
Mikkel Thorup
Lead Member of Technical Staff at AT&T Labs-Research
报告时间:
2007-06-07 14:00
报告地点:
FIT 4-603
主办单位:
理论计算机科学研究中心
  简介:

Mikkel Thorup is a Lead Member of Technical Staff at AT&T Labs-Research where  
he has been since 1998. He holds a PhD from Oxford University from 1993. From 1993 to 1998  
he was at the faculty of University of Copenhagen. Thorup's main work is in Algorithms and  
Data Structures and he is the editor of this area for Journal of the ACM. Currently he also  
serves on the editorial boards of SIAM Journal on Computing, ACM Transactions on Algorithms,  
and the open access journal Theory of Computing. Thorup has more than a hundred refereed  
publications and he is a co-inventor of the Smart Sampling Technologies that lie at the hart  
of AT&T's Scaleable Traffic Analysis Service.
        Thorup is a Fellow of the ACM, a Member of the Royal Danish Academy of Sciences, and  
an Adjunct Full Professor at the University of Copenhagen.
内容简介:We develop a new technique for proving cell-probe lower bounds for static data  
structures. Previous lower bounds used a reduction to communication games, which was known  
not to be tight by counting arguments. We give the first lower bound for an explicit problem  
which breaks this communication complexity barrier. In addition, our bounds give the first  
separation between polynomial and near linear space. Such a separation is inherently  
impossible by communication complexity. Using our lower bound technique and new upper bound  
constructions, we obtain tight bounds for searching predecessors among a static set of  
integers. We determine the optimal query time for any combination of space and word size w.  
In particular, we show that the classic van Emde Boas search time of O(log w) cannot be  
improved, even if we allow randomization. This is a separation from polynomial space, since  
Beame and Fich [STOC'99] give a predecessor search time of O(log w / log log w) using  
quadratic space.
 


 

今日相关信息
清华大学外语系学术讲座之一:The Villa...
聆听智者演绎人生、感悟为人治学之道
Regional & Urban Planning supported b...
压力应对有良方:谈大学生压力管理
 
同类别相关信息
清华信息大讲堂第141讲:数据质量管理的...
清华信息大讲堂第140讲:Big Data - Se...
Integrating and Accessing Multiling...
数据研究院RONG系列论坛一:大数据与新...
大数据论坛——数据科学与技术
学术活动