Difference between revisions of "Open Problems:50"

From Open Problems in Sublinear Algorithms
Jump to: navigation, search
(Created page with "{{Header |title=Sparse Recovery for Tree Models |source=bertinoro11 |who=Piotr Indyk }} For any $n=2^{h}-1$, we can identify the coordinates of a vector $v \in \mathbb R^n$ w...")
 
 
(5 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 
{{Header
 
{{Header
|title=Sparse Recovery for Tree Models
 
 
|source=bertinoro11
 
|source=bertinoro11
 
|who=Piotr Indyk
 
|who=Piotr Indyk
Line 16: Line 15:
 
'''Question:''' Is it possible to achieve $m=O(k)$ for some constant $C>0$?  
 
'''Question:''' Is it possible to achieve $m=O(k)$ for some constant $C>0$?  
  
'''Background:''' It is known that a weaker bound of $m=O(k \log(n/k))$ is possible to achieve even if ${\mathcal T}_k$ is replaced by the set of all $k$-subsets of $[n]$ {{cite|CandesRT-06a}}. However, since $|{\mathcal T}_k | =\exp(O(k))$, one can expect a better bound for ${\mathcal T}_k$. By using ''model-based compressive sensing'' framework of {{cite|BaraniukCDH-10}} (cf. {{cite|IndykP-11}}), one can achieve the desired bound of $m=O(k)$ but  with ''superconstant'' $C$.
+
'''Background:''' It is possible to achieve a weaker bound of $m=O(k \log(n/k))$ even if ${\mathcal T}_k$ is replaced by the set of all $k$-subsets of $[n]$ {{cite|CandesRT-06a}}. However, since $|{\mathcal T}_k | =\exp(O(k))$, one can expect a better bound for ${\mathcal T}_k$. The best $C$ we know how to achieve for $m = O(k)$ is $O(\sqrt{\log n})$ {{cite|IndykP-11}} (building on {{cite|BaraniukCDH-10}}).
 +
 
 +
==Update==
 +
Indyk and Razenshteyn {{cite|IndykR-13}} improved the bound on $m$ to $O(k \log(n / k) / \log \log (n / k))$ for constant $C$. In a follow-up paper, Bah et al. {{cite|BahBC-14}} presented an ''efficient'' recovery algorithm for this number of measurements.

Latest revision as of 22:15, 11 October 2013

Suggested by Piotr Indyk
Source Bertinoro 2011
Short link https://sublinear.info/50

For any $n=2^{h}-1$, we can identify the coordinates of a vector $v \in \mathbb R^n$ with the nodes of a full binary tree $B_h$ of height $h$ and root $1$. We define a $k$-sparse tree model ${\mathcal T}_k$ to be a set of all $T \subset [n]$ of size $k$ which form a connected subtree in $B_h$ rooted at $1$.

We want to design an $m \times n$ matrix $A$ such that for any $x \in \mathbb R^n$, one can recover from $Ax$ a vector $x^* \in \mathbb R^n$ such that:

\[ \left\|x^*-x\right\|_1 \leq \min_{x'\in\mathbb R^n,\ \operatorname{supp}(x') \subset T \mbox{ for some } T \in {\mathcal T}_k } C \left\|x'-x\right\|_1, \] where $\operatorname{supp}(x')$ is the set of non-zero coefficients of $x'$, and $C>0$ is a constant.

Question: Is it possible to achieve $m=O(k)$ for some constant $C>0$?

Background: It is possible to achieve a weaker bound of $m=O(k \log(n/k))$ even if ${\mathcal T}_k$ is replaced by the set of all $k$-subsets of $[n]$ [CandesRT-06a]. However, since $|{\mathcal T}_k | =\exp(O(k))$, one can expect a better bound for ${\mathcal T}_k$. The best $C$ we know how to achieve for $m = O(k)$ is $O(\sqrt{\log n})$ [IndykP-11] (building on [BaraniukCDH-10]).

Update[edit]

Indyk and Razenshteyn [IndykR-13] improved the bound on $m$ to $O(k \log(n / k) / \log \log (n / k))$ for constant $C$. In a follow-up paper, Bah et al. [BahBC-14] presented an efficient recovery algorithm for this number of measurements.