Amir Masoud Jafarpisheh

arXiv2026

Secure Multi-User Linearly-Separable Distributed Computing

The introduction of the new multi-user linearly-separable distributed computing framework, has recently revealed how a parallel treatment of users can yield large parallelization gains with relatively low computation and communication costs. These gains stem from a new approach that converts the computing problem into a sparse matrix factorization problem; a matrix $\mathbf{F}$ that describes the users' requests, is decomposed as $\mathbf{F} = \mathbf{DE}$, where a $γ$-sparse $\mathbf{E}$ defines the task allocation across $N$ servers, and a $δ$-sparse $\mathbf{D}$ defines the connectivity between $N$ servers and $K$ users as well as the decoding process. While this approach provides near-optimal performance, its linear nature has raised data secrecy concerns. We adopt an information-theoretic secrecy framework requiring that each user learns nothing more than its own requested function. Our main results provide (i) a necessary condition stating that for each user $k$ observing $α_k$ server responses, the common randomness visible to that user must span a subspace of dimension greater than $α_k-1$, and (ii) a necessary and sufficient condition requiring that removing from $\mathbf{D}$ the columns corresponding to the servers observed by a user leaves a matrix of rank at least $K-1$. Based on these conditions, we design a general, cost-preserving secrecy-enforcing transformation valid over both finite and real fields, obtained by appending to $\mathbf{E}$ a basis of $\mathrm{Null}(\mathbf{D})$ and carefully injecting shared randomness. This scheme preserves communication and computation costs, guarantees perfect information-theoretic secrecy over finite fields, and in the real case yields an explicit mutual-information bound that can be made arbitrarily small by increasing the variance of Gaussian common randomness.

cs.IT↗

arXiv2022

Distributed Attribute-based Private Access Control

In attribute-based access control, users with certain verified attributes will gain access to some particular data. Concerning with privacy of the users' attributes, we study the problem of distributed attribute-based private access control (DAPAC) with multiple authorities, where each authority will learn and verify only one of the attributes. To investigate its fundamental limits, we introduce an information theoretic DAPAC framework, with $N \in \mathbb{N}$, $N\geq 2$, replicated non-colluding servers (authorities) and some users. Each user has an attribute vector $\mathbf{v^*}=(v_1^*, ..., v_N^*)$ of dimension $N$ and is eligible to retrieve a message $W^{\mathbf{v}^*}$, available in all servers. Each server $n\in [N]$ is able to only observe and verify the $n$'th attribute of a user. In response, it sends a function of its data to the user. The system must satisfy the following conditions: (1) Correctness: the user with attribute vector $\mathbf{v^*}$ is able to retrieve his intended message $W^{\mathbf{v}^*}$ from the servers' response, (2) Data Secrecy: the user will not learn anything about the other messages, (3) Attribute Privacy: each Server~$n$ learns nothing beyond attribute $n$ of the user. The capacity of the DAPAC is defined as the ratio of the file size and the aggregated size of the responses, maximized over all feasible schemes. We obtain a lower bound on the capacity of this problem by proposing an achievable algorithm with rate $\frac{1}{2K}$, where $K$ is the size of the alphabet of each attribute.

cs.IT↗