Improved Algorithms for Adaptive Compressed Sensing

TL;DR

This paper advances adaptive compressed sensing algorithms, reducing measurement complexity and rounds for estimating sparse vectors, and extends improvements across various p-norm settings.

Contribution

It improves measurement bounds and adaptivity rounds for adaptive compressed sensing, and generalizes these improvements to all p, q norms with new bounds.

Findings

Reduced measurements to O(k log log(n/k)) for certain settings.

Achieved optimal O(log log n) rounds in adaptive schemes.

Extended improvements to all (p,p) norm cases for 0<p<2.

Abstract

In the problem of adaptive compressed sensing, one wants to estimate an approximately $k$-sparse vector $x\in\mathbb{R}^n$ from $m$ linear measurements $A_1 x, A_2 x,\ldots, A_m x$, where $A_i$ can be chosen based on the outcomes $A_1 x,\ldots, A_{i-1} x$ of previous measurements. The goal is to output a vector $\hat{x}$ for which $$\|x-\hat{x}\|_p \le C \cdot \min_{k\text{-sparse } x'} \|x-x'\|_q\,$$ with probability at least $2/3$, where $C > 0$ is an approximation factor. Indyk, Price and Woodruff (FOCS'11) gave an algorithm for $p=q=2$ for $C = 1+\epsilon$ with $\Oh((k/\epsilon) \loglog (n/k))$ measurements and $\Oh(\log^*(k) \loglog (n))$ rounds of adaptivity. We first improve their bounds, obtaining a scheme with $\Oh(k \cdot \loglog (n/k) +(k/\epsilon) \cdot \loglog(1/\epsilon))$ measurements and $\Oh(\log^*(k) \loglog (n))$ rounds, as well as a scheme with $\Oh((k/\epsilon) \cdot \loglog (n\log (n/k)))$ measurements and an optimal $\Oh(\loglog (n))$ rounds. We then provide novel adaptive compressed sensing schemes with improved bounds for $(p,p)$ for every $0 < p < 2$. We show that the improvement from $O(k \log(n/k))$ measurements to $O(k \log \log (n/k))$ measurements in the adaptive setting can persist with a better $\epsilon$-dependence for other values of $p$ and $q$. For example, when $(p,q) = (1,1)$, we obtain $O(\frac{k}{\sqrt{\epsilon}} \cdot \log \log n \log^3 (\frac{1}{\epsilon}))$ measurements.