from    
to    
search  

 


“Beyond the Diffraction Limit of Light....” Integrated Fluorescent-freeMult...
5·18国际博物馆日活动预告 | 刘骁:奇仪重器——探索科学博物馆的奥秘
环境学术沙龙第656期:Strengthening the Science-Policy Interface on Chemicals,W...
物理系colloquium: Quantum-enhanced metrology: theory and applications
报告题目:
Quadratic Lower Bounds on Matrix Rigidity
 报告人:
Prof Satya Lokam
Microsoft Research
报告时间:
2005-11-16 10:30
报告地点:
FIT楼1-222
主办单位:
姚期智教授组
  简介:

Title: Quadratic Lower Bounds on Matrix Rigidity

 

Speaker: Prof Satya Lokam  (Microsoft Research)

 

Place: Room 1-222, FIT Building         

Time: 10:30am, Nov 16

 

Abstract: The rigidity of a matrix $A$ with respect to the rank bound $r$ is the minimum number of entries of $A$ that must be changed to reduce the rank of $A$ to or below $r$. It is a major unsolved problem (Valiant, 1977) to construct ``explicit" families of $n \times n$ matrices of rigidity $n^{1+\delta}$ for $r=\epsilon n$ where $\epsilon$ and $\delta$ are positive constants. In fact, no superlinear lower bounds are known for explicit families of matrices for rank bound $r=\Omega(n)$.

 

We will present the first optimal, $\Omega(n^2)$, lower bound on the rigidity of two ``somewhat explicit" families of matrices with respect to the rank bound $r= cn$, where $c$ is an absolute positive constant. The entries of these matrix families are (i) square roots of the first $n^2$ primes and (ii) primitive roots of unity of prime orders for the first $n^2$ primes. Our proofs use an algebraic dimension concept introduced by Shoup and Smolensky (1997) and a generalization of that concept.

 

今日相关信息
超越材料性能的自然极限
 
同类别相关信息
Possible proximity of the Mott insu...
The multicore evolution and operati...
Price Differentiation in Communicat...
Game Theoretical Analysis of Vehicl...
The Future of Manycore Processors
学术活动