arXiv subjects

Foo Yee Yeo

Publications and source records attributed to Foo Yee Yeo.

3 recordsLinked to original sources

arXiv2023

Star-specific Key-homomorphic PRFs from Learning with Linear Regression

We introduce a novel method to derandomize the learning with errors (LWE) problem by generating deterministic yet sufficiently independent LWE instances that are constructed by using linear regression models, which are generated via (wireless) communication errors. We also introduce star-specific key-homomorphic (SSKH) pseudorandom functions (PRFs), which are defined by the respective sets of parties that construct them. We use our derandomized variant of LWE to construct a SSKH PRF family. The sets of parties constructing SSKH PRFs are arranged as star graphs with possibly shared vertices, i.e., the pairs of sets may have non-empty intersections. We reduce the security of our SSKH PRF family to the hardness of LWE. To establish the maximum number of SSKH PRFs that can be constructed -- by a set of parties -- in the presence of passive/active and external/internal adversaries, we prove several bounds on the size of maximally cover-free at most $t$-intersecting $k$-uniform family of sets $\mathcal{H}$, where the three properties are defined as: (i) $k$-uniform: $\forall A \in \mathcal{H}: |A| = k$, (ii) at most $t$-intersecting: $\forall A, B \in \mathcal{H}, B \neq A: |A \cap B| \leq t$, (iii) maximally cover-free: $\forall A \in \mathcal{H}: A \not\subseteq \bigcup\limits_{\substack{B \in \mathcal{H} \\ B \neq A}} B$. For the same purpose, we define and compute the mutual information between different linear regression hypotheses that are generated from overlapping training datasets.

cs.CR↗

arXiv2021

Function-private Conditional Disclosure of Secrets and Multi-evaluation Threshold Distributed Point Functions

Conditional disclosure of secrets (CDS) allows multiple parties to reveal a secret to a third party if and only if some pre-decided condition is satisfied. In this work, we bolster the privacy guarantees of CDS by introducing function-private CDS wherein the pre-decided condition is never revealed to the third party. We also derive a function secret sharing scheme from our function-private CDS solution. The second problem that we consider concerns threshold distributed point functions, which allow one to split a point function such that at least a threshold number of shares are required to evaluate it at any given input. We consider a setting wherein a point function is split among a set of parties such that multiple evaluations do not leak non-negligible information about it. Finally, we present a provably optimal procedure to perform threshold function secret sharing of any polynomial in a finite field.

cs.CR↗

arXiv2021

Extremal Set Theory and LWE Based Access Structure Hiding Verifiable Secret Sharing with Malicious-Majority and Free Verification

Secret sharing allows distributing a secret among several parties such that only authorized subsets, specified by an access structure, can reconstruct the secret. Sehrawat and Desmedt (COCOON 2020) introduced hidden access structures, that remain secret until some authorized subset of parties collaborate. However, their scheme assumes semi-honest parties and supports only restricted access structures. We address these shortcomings by constructing an access structure hiding verifiable secret sharing scheme that supports all monotone access structures. It is the first secret sharing scheme to support cheater identification and share verifiability in malicious-majority settings. The verification procedure of our scheme incurs no communication overhead. As the building blocks of our scheme, we introduce and construct: (i) a set-system with $> \exp\left(c\frac{2(\log h)^2}{(\log\log h)}\right)+2\exp\left(c\frac{(\log h)^2}{(\log\log h)}\right)$ subsets of a set of $h$ elements. Our set-system, $\mathcal{H}$, is defined over $\mathbb{Z}_m$, where $m$ is a non-prime-power. The size of each set in $\mathcal{H}$ is divisible by $m$ but the sizes of their pairwise intersections are not, unless one set is a subset of another, (ii) a new variant of the learning with errors (LWE) problem, called PRIM-LWE, wherein the secret matrix is sampled such that its determinant is a generator of $\mathbb{Z}_q^*$, where $q$ is the LWE modulus. The security of our scheme relies on the hardness of the LWE problem, and its share size is $$(1+ o(1)) \dfrac{2^{\ell}}{\sqrt{π\ell/2}}(2 q^{\varrho + 0.5} + \sqrt{q} + \mathrmΘ(h)),$$ where $\varrho \leq 1$ is a constant and $\ell$ is the total number of parties. We also provide directions for future work to reduce the share size to \[\leq \dfrac{1}{3} \left( (1+ o(1)) \dfrac{2^{\ell}}{\sqrt{π\ell/2}}(2 q^{\varrho + 0.5} + 2\sqrt{q}) \right).\]

cs.CR↗