Abstract:In the subspace sketch problem one is given an $n\times d$ matrix $A$ with $O(\log(nd))$ bit entries, and would like to compress it in an arbitrary way to build a small space data structure $Q_p$, so that for any given $x \in \mathbb{R}^d$, with probability at least $2/3$, one has $Q_p(x)=(1\pm\epsilon)\|Ax\|_p$, where $p\geq 0$, and where the randomness is over the construction of $Q_p$. The central question is: How many bits are necessary to store $Q_p$? This problem has applications to the communication of approximating the number of non-zeros in a matrix product, the size of coresets in projective clustering, the memory of streaming algorithms for regression in the row-update model, and embedding subspaces of $L_p$ in functional analysis. A major open question is the dependence on the approximation factor $\epsilon$. We show if $p\geq 0$ is not a positive even integer and $d=\Omega(\log(1/\epsilon))$, then $\tilde{\Omega}(\epsilon^{-2}d)$ bits are necessary. On the other hand, if $p$ is a positive even integer, then there is an upper bound of $O(d^p\log(nd))$ bits independent of $\epsilon$. Our results are optimal up to logarithmic factors, and show in particular that one cannot compress $A$ to $O(d)$ "directions" $v_1,\dots,v_{O(d)}$, such that for any $x$, $\|Ax\|_1$ can be well-approximated from $\langle v_1,x\rangle,\dots,\langle v_{O(d)},x\rangle$. Our lower bound rules out arbitrary functions of these inner products (and in fact arbitrary data structures built from $A$), and thus rules out the possibility of a singular value decomposition for $\ell_1$ in a very strong sense. Indeed, as $\epsilon\to 0$, for $p = 1$ the space complexity becomes arbitrarily large, while for $p = 2$ it is at most $O(d^2 \log(nd))$. As corollaries of our main lower bound, we obtain new lower bounds for a wide range of applications, including the above, which in many cases are optimal.

Tighter bound for generalized multiple discrete logarithm problem via MDS matrix method

Hard Instances of the Constrained Discrete Logarithm Problem

Exact Parameterized Multilinear Monomial Counting via k-Layer Subset Convolution and k-Disjoint Sum.

On the Construction of Near-MDS Matrices

New Public-Key Cryptosystem Based on Two-Dimension DLP

Polynomial Integrality Gaps for Strong SDP Relaxations of Densest K-Subgraph

An Exact Algorithm with New Upper Bounds for the Maximum k-Defective Clique Problem in Massive Sparse Graphs

Tight Dynamic Problem Lower Bounds from Generalized BMM and OMv

On the Direct Construction of MDS and Near-MDS Matrices

Probability Method for Cryptanalysis of General Multivariate Modular Linear Equation

A quasi-polynomial time algorithm for Multi-Dimensional Scaling via LP hierarchies

Information-Theoretically Private Matrix Multiplication From MDS-Coded Storage

On the success probability of the quantum algorithm for the short DLP

On the Complexity of Generalized Discrete Logarithm Problem

Tight finite-key analysis for generalized high-dimensional quantum key distribution

Improved lower bound for Diffie–Hellman problem using multiplicative group of a finite field as auxiliary group

Tight Bounds for the Subspace Sketch Problem with Applications

Block-MDS QC-LDPC Codes for Information Reconciliation in Key Distribution

Constructing linear transformations of MPKC by generalized central symmetric matrices

Bipartite Matching in Nearly-linear Time on Moderately Dense Graphs

Generalized GM-MDS: Polynomial Codes are Higher Order MDS